A stable partitioned FSI algorithm for rigid bodies and incompressible flow. Part I: Model problem analysis

Qi Tang, Jeffrey W. Banks, William D. Henshaw, Donald W. Schwendeman, Qi Tang 0001

Introduction

Simulating the interaction between fluids and solid structures (Fluid-Structure Interaction, or FSI) is a cornerstone of modern engineering, essential for designing everything from heart valves to aircraft wings. However, researchers have long struggled with the "added-mass" instability. This numerical phenomenon occurs when a solid body is light relative to the fluid it displaces; the fluid acts as an additional inertial weight that standard partitioned algorithms fail to account for, leading to catastrophic numerical oscillations.

Schematic of a rigid disk in a fluid domain, illustrating translation (left) and rotation (right) along with the viscous diffusion length $\Delta n$.

Traditional partitioned schemes, which solve fluid and solid equations separately for modularity, typically require many expensive sub-iterations per time step to remain stable in these "light-body" scenarios. This paper introduces the Added-Mass Partitioned for Rigid Bodies (AMP-RB) algorithm, a breakthrough that achieves unconditional stability without the need for sub-iterations.

The AMP-RB Algorithm

The core innovation of the AMP-RB scheme is the use of a generalized Robin interface condition. Unlike standard methods that simply pass velocity and stress back and forth between solvers, AMP derives a boundary condition for the fluid pressure that explicitly incorporates the linear and angular acceleration of the rigid body. This effectively "bakes" the added-mass effect directly into the fluid solver's pressure step.

To handle viscous instabilities—often referred to as "added-damping"—the authors introduce added-damping tensors. These are derived through a linearization of force and torque integrals, allowing the algorithm to anticipate how viscous shear will affect the body's motion. A key discovery in the methodology is the introduction of a scaling parameter, $\beta_d$. The authors found that setting $\beta_d \approx 1$ provides a robust "sweet spot" for stability across a wide range of physical parameters. For extremely light bodies, an optional Velocity Correction (VC) step further ensures stability against viscous drag.

Stability Analysis and Model Problems

To prove the algorithm's robustness, the authors performed rigorous mathematical analysis using Fourier and GKS stability theory. They focused on simplified model problems that capture the essential physics of the FSI challenge, such as 1D piston motion and 2D disks in annular geometries.

The rectangular model problem geometry, showing the fluid domain $\Omega$, the rigid body $\Omega_b$, and the interface $\Gamma_b$.

The analysis demonstrates that while traditional schemes are only stable if the body's mass exceeds the fluid's added mass, the AMP scheme remains stable even when the body mass is zero. This theoretical foundation was verified using several benchmarks, including newly derived exact solutions for translating disks in annular Stokes flow.

Key Results

  • Unconditional Stability: AMP-RB is shown to be stable for any mass ratio, effectively solving the added-mass and added-damping instability problems.
  • Second-Order Accuracy: Numerical tests confirm that the scheme maintains second-order accuracy in both space and time, with errors dropping quadratically as the grid is refined.
  • Computational Efficiency: By eliminating the need for sub-iterations, the AMP scheme is significantly faster than traditional methods, which might otherwise require 10–100 iterations per time step to maintain stability for light bodies.

Conclusion and Future Work

This work establishes a stable and efficient framework for FSI simulations involving rigid bodies. By addressing the fundamental mathematical causes of instability, the AMP-RB algorithm allows for high-fidelity simulations of light or even massless structures in incompressible flow. While this first part focuses on rigid bodies and simplified geometries, the authors plan to extend these principles to general, complex geometries and deforming elastic structures in future research.

Summary updated

Abstract

A stable partitioned algorithm is developed for fluid-structure interaction (FSI) problems involving viscous incompressible flow and rigid bodies. This {\em added-mass partitioned} (AMP) algorithm remains stable, without sub-iterations, for light and even zero mass rigid bodies when added-mass and viscous added-damping effects are large. The scheme is based on a generalized Robin interface condition for the fluid pressure that includes terms involving the linear acceleration and angular acceleration of the rigid body. Added-mass effects are handled in the Robin condition by inclusion of a boundary integral term that depends on the pressure. Added-damping effects due to the viscous shear forces on the body are treated by inclusion of added-damping tensors that are derived through a linearization of the integrals defining the force and torque. Added-damping effects may be important at low Reynolds number, or, for example, in the case of a rotating cylinder or rotating sphere when the rotational moments of inertia are small. In this first part of a two-part series, the properties of the AMP scheme are motivated and evaluated through the development and analysis of some model problems. The analysis shows when and why the traditional partitioned scheme becomes unstable due to either added-mass or added-damping effects. The analysis also identifies the proper form of the added-damping which depends on the discrete time-step and the grid-spacing normal to the rigid body. The results of the analysis are confirmed with numerical simulations that also demonstrate a second-order accurate implementation of the AMP scheme.
BibTeX
@article{tang2017stable,
  author = {Qi Tang and Jeffrey W. Banks and William D. Henshaw and Donald W. Schwendeman and Qi Tang},
  title = {{A stable partitioned FSI algorithm for rigid bodies and incompressible flow. Part I: Model problem analysis}},
  journal = {Journal of Computational Physics},
  year = {2017},
  doi = {10.1016/j.jcp.2017.01.015},
}
Figure