Black_Banner_New.jpg

AERO-F

AERO-F is a three-dimensional, domain-decomposition-based, compressible Euler/Navier-Stokes solver with turbulence modeling capabilities. It can simulate single-material and highly nonlinear multi-material flow problems involving arbitrary structural systems. To this effect, AERO-F can model a fluid as a perfect or pseudo-incompressible gas (possibly going through porous media and/or subjected to a discrete gust wind), a stiffened gas, a barotropic liquid or gas governed by Tait's Equation of State (EOS), a medium governed by Tillotson's EOS, or air at thermochemical equilibrium. Using various level set techniques, it is capable of capturing arbitrary material interfaces, including fluid-structure interfaces undergoing large motions and/or deformations and crack propagation.

AERO-F performs steady-state and unsteady, inviscid (Euler) and viscous (Navier-Stokes) laminar as well as turbulent flow simulations past still or moving, accelerating or decelerating, rigid or flexible obstacles. For turbulent flow computations, it offers one- and two-equation turbulence models, static and dynamic LES (Large Eddy Simulations) and VMS-LES (Variational Multi-Scale LES), as well as DES (Detached Eddy Simulation) methods, with or without a wall function. For flexible obstacles, it communicates with the structural/thermal analyzer AERO-S to perform high-fidelity, high-accuracy, aeroelastic, gust response, and other types of fluid-structure analyses. AERO-F can also perform high-fidelity fluid-thermal (aerothermal) and fluid-thermal-structure (aerothermoelastic) analyses by appropriately communicating with AERO-S.

AERO-F also offers a two-step approach for performing aeroacoustic and hydroacoustic computations that is equipped with two different acoustic analogies: Kirchhoff's integral method, and the Morfey-Wright acoustic analogy which generalizes the density-based Ffowcs Williams-Hawkings acoustic analogy to a pressure formulation. In this two-step approach, the shape and position of the Kirchhoff surface can be specified by the user. Such a surface can be rigid, flexible, stationary, or moving. The ambient fluid can be at rest, or moving at a uniform free-stream velocity. This aeroacoustic or hydroacoustic capability is fully integrated with the other simulation capabilities of AERO Suite, including those pertaining to coupled aeroelastic problems.

AERO-F operates on unstructured body-fitted meshes that can combine tetrahedra, prisms, pyramids, and hexahedra. It can also compute flows past and/or within static and dynamic obstacles whose wet surfaces are embedded in a fixed, tetrahedral or hexahedral, non- body-fitted, embedding mesh.  The body-fitted meshes and the embedded discrete surfaces can be fixed, moved and/or deformed in a prescribed manner (forced body motion), or via the interaction with the structural code AERO-S. In the case of body-fitted meshes, the governing equations of fluid motion are formulated in the arbitrary Lagrangian Eulerian (ALE) framework. In this case, large mesh motions can be handled by a corotational approach that separates the rigid and deformational components of the motion at the surface of the obstacle, sliding planes or surfaces that prevent mesh shearing, and robust mesh motion algorithms that are based on structural analogies. In the case of embedded surfaces, which can have complex shapes and arbitrary thicknesses, the governing equations of fluid motion are formulated in the Eulerian framework and the wall boundary or transmission conditions are treated by an embedded boundary method.

AERO-F also features a Chimera (overset grid) capability that supports most of its functionalities, including dynamic and deformable ALE body-fitted meshes.

The spatial discretization adopted by AERO-F blends the finite volume and finite element methods. More specifically, this semi-discretization combines a second-order accurate Roe, HLLE, or HLLC upwind scheme for the convective fluxes and a Galerkin centered approximation of the viscous fluxes. It can achieve a fifth-order spatial dissipation error and a sixth-order spatial dispersion error — and therefore fifth-order spatial accuracy. In some specific circumstances, it can also achieve sixth-order spatial accuracy. Time-integration in AERO-F can be performed with first- and second-order implicit, and first-, second-, and fourth-order explicit algorithms that satisfy their respective discrete geometric conservation laws.

AERO-F embeds AERO-FL, a linearized flow solver sharing with AERO-F the semi-discretization schemes outlined above. This linearized module can be used to compute flow perturbations around an equilibrium solution, construct a set of generalized aerodynamic and/or aerodynamic force matrices, predict linearized aeroelastic, aerothermal, and aerothermoelastic responses assuming a modalized structure and a thermal sub-system represented by a Proper Orthogonal Decomposition (POD)-based Reduced-Order Model (ROM), compute fluid snapshots in either the time- or frequency-domain to construct a fluid POD basis, interpolate a set of previously computed fluid POD bases, generate a fluid, aeroelastic, aerothermal, or aerothermoelastic ROM, and compute fluid, aeroelastic, aerothermal, and aerothermoelastic ROM solutions in the time-domain assuming a modalized structure and a thermal sub-system represented by a POD-based ROM.

AERO-F can also perform steady-state, inviscid and viscous sensitivity analyses around a specified steady-state flow solution with respect to a specified set of flow and shape parameters.

In summary, AERO-F can perform on its own the following computations or analyses:

Complex Flow Analysis

  • Steady-state or unsteady flow computations around a fixed rigid obstacle.
  • Steady-state or unsteady natural convection (buoyancy) computations.
  • Unsteady flow computations around a rigid or flexible obstacle set in accelerated or decelerated motion.
  • Unsteady flow computations around a rigid or flexible body set into a prescribed motion or deformation (forced body motion), and/or with a prescribed oscillatory wall boundary temperature (forced temperature oscillations).
  • Unsteady flow computations around an accelerated or decelerated rigid or flexible body set into a prescribed motion or deformation (forced body motion), and/or with a prescribed oscillatory wall boundary temperature (forced temperature oscillations).
  • Any of the above computations in the presence of porous media.
  • Unsteady flow computations around a rigid or flexible obstacle in a landing configuration, rectilinearly rolling out on the runway, with tire friction, braking forces, and possibly landing gear dynamics. 

Trimming Analysis

  • Steady-state trim computations for a rigid vehicle equipped or not with deployable control surfaces. 

Maneuvering Analysis

  • Unsteady flow computations around a maneuvering rigid vehicle with dynamic control surfaces and control laws.

Gust Analysis

  • Any of the above unsteady computations in the presence of a discrete gust wind. 

Linearized Complex Flow Analysis

  • Unsteady linearized and perturbed flow computations in the time domain.
  • Unsteady linearized aeroelastic computations where the structure is represented by a truncated set of its natural modes.
  • Unsteady linearized and perturbed fluid-thermal (aerothermal) computations in the time domain where the thermal field in the structure is represented by a ROM.
  • Unsteady linearized and perturbed fluid-thermal-structure (aerothermoelastic) computations in the time domain, where the thermal field in the structure is represented by a reduced-order model, and the structure itself is represented by a truncated modal superposition. 

Model Order Reduction and Linearized Multi-Physics Analysis using Reduced-Order Models

  • Construction of a time- or frequency-domain Proper Orthogonal Decomposition (POD) basis trained for harmonic obstacle vibrations and/or wall boundary temperature oscillations.
  • Construction of a time- or frequency-domain POD basis by interpolation of given sets of POD basis vectors.
  • Construction of a generalized aerodynamic and/or aerodynamic force matrix (or a set of such matrices).
  • Construction of a fluid ROM trained for harmonic obstacle vibrations and/or wall boundary temperature oscillations.
  • Time-domain fluid ROM computations where the flow is expressed in a POD basis.
  • Construction of an aeroelastic ROM trained for harmonic obstacle vibrations and/or wall boundary temperature oscillations where the fluid is represented by a POD basis, and the structure is represented by a truncated set of its natural modes.
  • Time-domain aeroelastic ROM computations where the structure is represented by a truncated set of its natural modes.
  • Construction of a fluid-thermal (aerothermal) ROM trained for harmonic wall boundary temperature oscillations, where the viscous fluid is represented in a POD basis, and the thermal field in the structure is represented in a different POD basis.
  • Time-domain fluid-thermal (aerothermal) ROM computations where the fluid is expressed in a POD basis, and the thermal field in the structure is represented in a different POD basis.
  • Construction of an aerothermoelastic ROM trained for harmonic obstacle vibrations and wall boundary temperature oscillations, where the fluid is represented in a POD basis, the thermal field in the structure is represented by a POD-based ROM, and the structure itself is represented by a truncated modal superposition.
  • Time-domain aerothermoelastic ROM computations, where the fluid is expressed in a POD basis, the thermal field in the structure is represented in a different POD basis, and the structure itself is represented by a truncated modal superposition. 

Multi-Fluid or Multi-Phase Flow Analysis

  • Steady-state or unsteady, highly nonlinear, multi-fluid or multi-phase flow analysis. 

Sensitivity Analysis

  • Steady-state sensitivity analysis around a specified steady-state flow solution with respect to a specified set of flow and shape parameters. 

Aeroacoustic Analysis

  • Frequency-domain computation of the acoustic pressure and density radiated from a rigid obstacle, their power spectral densities in the far-field at user-specified locations, and their far-field patterns when the free-stream velocity is zero.
  • Time-domain computation of the acoustic pressure and density radiated from a rigid obstacle in the far-field at user-specified locations.