Approaches to Industrial Fluid-Structures Interaction
Fred Mendonça, CD-adapco

Introduction
Virtual Product Development (VPD) continues to place demands on CAE analysis including multi-disciplinary activities such as fluid-structure interaction. FSI covers a wide range of applications; some of the more promising simulation methodologies are:

Parallel execution of single-discipline software, such as CFD and FEA, communicating relevant information during execution via a controller such as MpCCI [1]. Capable of dealing with the full non-linearities in both fluid and structure systems, the coupling is performed explicitly.

Single execution of mutli-discipline software, in which the finite-volume or finite-element methodology solves simultaneously both the fluid dynamics and solid mechanics [2]. Here, the full non-linearities in both fluid and structure systems are implicitly coupled, and is able also to handle casting, melting and solidification.

Substructure matrix approach in which the mass, stiffness and damping matrices for the structure are exported from the structures code to the fluids code and inverted through an API on the basis of the fluid loading. This methodology is limited to linear deformations.

Δ Figure 2: Instantaneous VOF shortly after striking the surface water sheet (left) and after the flow conditions become pseudo-steady (right).

6-DOF (six degrees of freedom) systems in which the deformations may or may not be rigid, whilst the structure is allowed to rotate and translate, e.g. floating bodies and store release.

In this article illustrates several examples, covering different sectors of industry including automotive, power generation, nuclear engineering, bio-medical, marine engineering, oil and gas, with applications ranging from tube-bundle vortex induced vibration (VIV), automotive exhaust system thermal stressing and tyre aqua/hydro-planing, to floating bodies.

Examples
MpCCI approach
This is a very general coupling simulation methodology, which permits established and consequently well-validated, single-discipline codes to be coupled together for multi-physics applications. STAR-CD interfaces to all structures codes supported by MpCCI, including ABAQUS, ANSYS and PERMAS.

In the thermal analysis of a 4-cylinder twin-catalyst exhaust system, the metal components are exposed to temperatures which are beyond linear elastic limits, and are therefore susceptible to plastic deformation over many heat-up/cool-down cycles. A coupled STAR-CD and MSC.Marc transient heat transfer analysis predicts the equivalent plastic strain distribution in the metal skin. Large plastic deformations are found at the exterior surface near the four support points on both sides of the catalysts.

Tyre contact with the road in the presence of surface water affects the traction and therefore dynamic stability of the vehicle. The major challenge to the modelling of hydro- or aqua-planing is to combine the non-trivial handling of the complicated treaded tyre geometry (simple longitudinal as well as complex grooves) with deformation caused by the forces induced in a two-phase water/air system whilst the tyre is rolling.

A faithful representation is achieved by using STAR-CD’s free surface capability together with dynamic motion / sliding mesh capability together with an efficient transient solution methodology, then communicating the fluid forces to a rubber deformation model of the tyre, which in turn returns the deformation to the fluid mesh.

Finite Volume combined fluid and structural approach Computational Continuum Mechanics (CCM) is the termed coined for a combined approach to multi-physics such as fluid and structural mechanics.As described in [3] the conservation equations apply to both and may be solved using a finite volume methodology whereas the constitutive relations and material properties differ between the two systems. CCM for FSI, enabled in STAR-CD V4, offers some advantages compared to traditional
FEA methodologies. It is applicable to a wider range of applications including casting, solidification, melting and floating bodies, but becomes potentially unstable for structural analysis of thin structures (plates/shells).

Our CCM implementation offers the added benefits of fully automated polyhedral meshing and full HPC-parallel scalability. The manoeuvring effectiveness and structural integrity of rudders in crossflow, typical in marine engineering, in the wake of the ships hull and in vicinity of the propeller(s) may be studies through this approach.

The figures illustrate fluid velocities near/behind the rudder and also the pressurisation on the inclined faces, which result in structural stresses and displacement.

Click image (left) to enlarge

Slamming of ship structures in rough sea-weather conditions imposes severe structural loading on the hull. The two drop-tests described below, for a hollow cylinder and a solid inclined wedge, predict the motion due to gravity and interaction with the free-surface flow, leading to wall deformations and 6DOF motion.

Full 6DOF motion results from the inclined wedge drop-test – here the computed linear and angular acceleration, together with the heel angle and impact velocity, are compared with experiment [4].

Substructure matrix approach This is a unique approach in which any wellestablished structures code may be used to generate the mass, stiffness and damping matrices, then exported to files. CD-adapco’s expert system, es-fsi, reads in the matrix file and solves the dynamics equations for the structural displacements within the CFD solutions system. Substructuring is used to condense out many degrees of freedom with respect to the fluid mesh. The fluid forces computed by STAR-CD are interpolated onto the substructure nodes and the matrix inverted in a special-purpose API to compute the displacements at each time-step. The coupling
method is explicit, but benefits by eliminating all communication overheads.

Vortex-induced vibration (VIV) is common in heat-exchanger applications in the power generation, oil and gas, process and nuclear industries. In this example, the flow past a tube array creates a vortex shedding pattern, which perturbs the tubes. The perturbations result eventually in self-sustaining oscillations. The limiting amplitude of oscillation increases with the oncoming flow velocity, until at some critical flow velocity, in this case 1.2m/s, the amplitude exceeds the separation between tubes, causing them to collide.