Pieces are coming together for CFD Vision 2030

The CFD Vision 2030 Integration Committee advocates for, inspires and enables community activities recommended by the vision study for revolutionary advances in the state-of-the-art of computational technologies needed for analysis, design and certification of future aerospace systems.

Demonstrating efficiently scaled computational fluid dynamics simulations on an exascale system is a key technology milestone for the CFD Vision 2030 activity. The term “exascale” refers to a system capable of performing one exaflop, or 1018 floating-point operations per second. The Frontier system at the Oak Ridge National Laboratory in Tennessee passed acceptance testing in April, making it available to over 1,000 developers and researchers worldwide. GE Aerospace and the FUN3D group at NASA’s Langley Research Center in Virginia are among the early users of this world-leading system, using it to attempt simulations that would have been impossible on previous systems. GE is using Frontier to assess aircraft engine design trade-offs, from the microscopic features of turbulence to the larger effects of how these flow features evolve and affect system performance and noise. NASA Langley researchers are pursuing simulations of long-duration reentry trajectories into the Martian atmosphere using retropropulsion to meet this key 2024 technology milestone. Located at the Argonne National Laboratory in Illinois, a second exascale-class system known as Aurora reached an important milestone in June when hardware installation was completed. Aurora is projected to provide a computational capacity exceeding two exaflops.

Improvements in computer speed and memory size are key enablers for simulating the complex flow fields around aircraft and spacecraft. To leverage these capabilities efficiently, improvements are needed for solver algorithms and post processing. The AIAA CFD High Lift Prediction workshops have included a growing number of higher-order solvers, highlighting the need for higher-order meshing and visualization. Fortunately, commercial developers are developing these capabilities. Pointwise from California-based Cadence Design Systems continues to improve its higher-order meshing, and Washington-based Tecplot Inc. released a new version of its Tecplot visualization code in August that includes support for high-order mesh elements.

Launchpad damage after the SpaceX Starship launch in April and the Artemis I Space Launch System launch in 2022 showed the harsh environments that launchpads must endure and the challenges associated with protecting them. Full-scale testing and analysis of these dynamic, chemically reacting, multiphase environments are major challenges. The Launch, Ascent, and Vehicle Aerodynamics, LAVA, group at NASA’s Ames Research Center in California is working with NASA’s Exploration Ground Systems in Florida to validate its code with the Artemis I flight data and develop new environments for Artemis II using tools that require high-performance computing, complex physical models and robust algorithms to understand these environments.

A key goal of AIAA’s CFD Vision 2030 Integration Committee is to see leading-edge research moved into production codes that can benefit industry and government programs. An example of such progress is Bell’s V-280 tiltrotor. Its selection last year as the U.S. Army’s Future-Long Range Assault Aircraft was upheld in April by the U.S. Government Accountability Office after a bid protest by Sikorsky, whose coaxial rotor aircraft, the Defiant X, was turned down by the Army. The decision to select the V-280 for the Army’s next major aircraft development program was informed in part by high-fidelity CFD/computational structural dynamics calculations the Army performed on both configurations. Specifically, since 2014 the Army Combat Capabilities Development Command Aviation and Missile Center has investigated the performance of both vehicles by performing calculations on the Defense Department’s High Performance Computing Modernization Program’s massively parallel computer systems using the Computational Research and Engineering Acquisition Tools and Environments Air Vehicle, or CREATE-AV, Helios software. Scaled-model tests performed by the companies validated the accuracy of the high-fidelity models.

Pieces are coming together for CFD Vision 2030