You are here

Modern, Rapid, Usable, Lower Order Computational Fluid Dynamics (CFD) Development for Aerodynamic Analysis

Description:

TECHNOLOGY AREA(S): Air Platform 

OBJECTIVE: Develop a rapid and easy to use lower order computational fluid dynamics method to analyze the aerodynamic performance, stability, and control of air vehicles in order to aid design, simulation, and technology assessment. 

DESCRIPTION: AFRL’s Aerodynamic Technology branch is tasked with discovering, developing, and demonstrating aerodynamic technologies for war-winning capabilities for the USAF. A range of technical activities including configuration aerodynamics, flow control device development, high fidelity computational tool development and application, and wind tunnel testing must be conducted to reach these goals. The accessibility of multi-core computers, Reynolds Averaged Navier Stokes codes, and aircraft geometry modeling tools has made the use of high fidelity CFD much more common; however, the time required for computational execution and processing 'touch' labor by the engineer is still much greater for RANS analyses compared to lower order methods. While some flow regimes require higher fidelity, often a reduced order method can suitably model vehicle aerodynamics and speed up design or analysis processes. These lower order methods offer attractive tradeoffs between computational solution time and accuracy, and with modern software approaches, numerical schemes, and other simulation techniques, be beneficial for use in early aerodynamic predictions and design. While using reduced order physics equations to compute accurate, detailed pressure fields on and off the vehicle body efficiently, geometry is still represented in full 3D definition. These results can identify drag contributors, develop structural loads, and compute stability and control derivatives, and serve a host of other aeronautical engineering purposes. Once a vehicle has been designed by a lower order method, it can be passed along to Euler or RANS methods to develop the design further (as necessary). Though potential methods have been in use for 50 years, they remain key tools in today's multi-fidelity aerodynamic analysis environment. However, there is much that can be leveraged from recent software, numerical, and computational hardware advances to improve ease-of-use, speed, and accuracy of reduced order methods. Desired attributes of a modern lower order aero tool include ease-of-use, speed, and accuracy. Additional considerations and capabilities are necessary so the code is compatible with today's analysis processes. Ease-of-use: the tool should be able to: be used on multiple OSs; take unordered quad panel or triangulated surface meshes as input; not require specification of wake geometry; be equipped with a GUI geometry viewer to aid in setup and processing of the analysis; be compatible with a variety of input formats. Speed: the tool should: use modern numerical techniques to rapidly and efficiently solve large mesh solutions; utilize a specifiable number of processors; offer ease-of-use attributes to reduce 'hands on' time required to setup the cases and post-process the results - in doing so, overall end-to-end solution time is reduced. Accuracy: the tool should be able to: handle suitably large, dense surface meshes; handle multi-body meshes; model propulsion related aerodynamic effects; deform/relax the wake mesh to capture vortex and wake effects; estimate viscous drag by way of on-body streamline tracing or strip-wise computations; estimate or correct for compressibility effects, as appropriate. 

PHASE I: Demonstrate the feasibility of the proposed analysis tool to easily and quickly evaluate aerodynamic performance. Initial system architecture and proposed interfaces will be defined, and core algorithms prototyped. An alpha version of the tool should be produced (not full featured). 

PHASE II: Continued development of the tool, to include implementation and refinement of the core features, and the user interface and viewer. Iterative beta releases should be delivered as tool matures. Draft user's manual and theory document should be produced. Validation of aerodynamic results against trusted data to be shown. Sample integration with CAD/geometry tool derived inputs and shape design code. Exercise of all required features. Maturing commercialization plan. 

PHASE III: Continued refinement of the tool, to include support for multiple operating systems, licensing schemes, and expanding interface to aid in speed and ease of use. 

REFERENCES: 

1. Willis, D., Peraire, J., and White, J., "A Combined pFFT - Multipole Tree Code, Unsteady Panel Method with Vortex Particle Wakes", AIAA-2005-0854; 2005.; 2. P. A., Henne (editor); "Applied Computational Aerodynamics – Progress in Astronautics and Aeronautics", Volume 125 - AIAA; 1990.; 3. Calabretta J., and McDonald R., "A Three Dimensional Vortex Particle Panel Method for Modeling Propulsion Airframe Interaction", AIAA-2010-0679; 2010.; 4. Willis, D., "Enriched Basis Functions for Automatically Handling Wake-Body Intersections in Source-Doublet-Potential Panel Methods", AIAA-2012-0265; 2012.

KEYWORDS: Lower Order CFD, Potential Method, Panel Method, Aerodynamics, Software Development, Fast Fourier Transform, Subsonic Vehicle Design, Aerodynamic Design, Propulsion Integration 

US Flag An Official Website of the United States Government