An efficient scalable framework for aeroelastic analysis and adjoint-based sensitivities using FUN3D and tacs next-generation aircraft designs, such as NASA's n+3 concepts, exhibit greater flexibility…

This federal contract award, NNX15AU22A, was granted to Georgia Tech Research Corporation by the National Aeronautics and Space Administration (NASA). The contract, valued at $998,969.94, is for the development of an "Efficient Scalable Framework for Aeroelastic Analysis and Adjoint-Based Sensitivities Using FUN3D and TACS" to support the analysis and design of next-generation flexible aircraft, such as NASA's N+3 concepts.

The contract integrates NASA's Computational Fluid Dynamics (CFD) code FUN3D and Georgia Tech's Toolkit for the Analysis of Composite Structures (TACS) to create an aeroelastic analysis and adjoint-based sensitivity tool called FUNTOFEM. This framework will leverage the existing capabilities of both tools and extend them to achieve high-fidelity aeroelastic analysis and design optimization for flexible aircraft. The contract does not have a set-aside designation.

An efficient scalable framework for aeroelastic analysis and adjoint-based sensitivities using FUN3D and tacs next-generation aircraft designs, such as NASA's n+3 concepts, exhibit greater flexibility driven by advanced structural technologies and ever-more demanding fuel burn goals. As a result of this flexibility, flutter and adverse aeroelastic phenomena are becoming increasingly important design requirements that must be considered early in the design process to avoid excessive weight penalties. Furthermore, due to the large deflections exhibited by these aircraft under static and transient aerodynamic loads, linear structural analysis is no longer representative and nonlinear structural analysis must be employed. To meet the challenges of analyzing and designing the next generation of flexible aircraft, we propose to develop a high-fidelity aeroelastic analysis and adjoint-based sensitivity tool. Our proposed framework will integrate the capabilities of the NASA CFD code FUN3D, with the toolkit for the analysis of composite structures (tacs), a parallel finite-element code with nonlinear time-dependent structural analysis capability. This aeroelastic tool, called funtofem, will leverage the existing capabilities of both tacs and FUN3D and will extend these capabilities in order to achieve an integrated aeroelastic analysis and adjoint-based sensitivity framework with excellent parallel performance. In particular, we will use the existing aerodynamic adjoint implementation in FUN3D, including elasticity-based mesh motion methods, and the accurate and efficient structural adjoint method implemented in tacs which has been verified using complexstep comparisons. To couple these codes we will use a modular approach that will offer both loosely-coupled and tightly-coupled capabilities. Time-dependent analysis and adjoint methods must be implemented with care to achieve good parallel scalability. Therefore, we will develop methods that will partition the CFD and structural meshes for optimal communication patterns when performing aeroelastic analysis and adjoint calculations. The computational cost of an efficient adjoint implementation should scale very weakly with increasing number of aerodynamic, geometric and structural parameters. To achieve this goal, we will implement methods with an integrated mesh-adjoint to avoid multiple mesh-motion operations at each aeroelastic adjoint time-step. We will demonstrate our framework on flexible high aspect ratio aircraft including the truss-braced wing and high altitude long endurance aircraft (hale).