FluSI: A novel parallel simulation tool for flapping insect flight using a Fourier method with volume penalization (1506.06513v3)

Abstract: FluSI, a fully parallel open source software for pseudo-spectral simulations of three-dimensional flapping flight in viscous flows, is presented. It is freely available for non-commercial use under [https://github.com/pseudospectators/FLUSI]. The computational framework runs on high performance computers with distributed memory architectures. The discretization of the three-dimensional incompressible Navier--Stokes equations is based on a Fourier pseudo-spectral method with adaptive time stepping. The complex time varying geometry of insects with rigid flapping wings is handled with the volume penalization method. The modules characterizing the insect geometry, the flight mechanics and the wing kinematics are described. Validation tests for different benchmarks illustrate the efficiency and precision of the approach. Finally, computations of a model insect in the turbulent regime demonstrate the versatility of the software.

• The paper introduces FluSI, a novel open-source parallel simulation tool for accurately modeling the complex aerodynamics of flapping insect flight.
• FluSI employs a Fourier pseudo-spectral method with volume penalization to handle dynamic geometries and viscous flows on distributed memory HPC systems using MPI.
• Validation against known cases, including hovering fruit flies, demonstrates FluSI's ability to simulate vortical structures and inertial forces critical for bio-inspired aerodynamic research.

Overview of FluSI: A Parallel Simulation Tool for Flapping Insect Flight

This paper introduces FluSI, an open-source computational framework specifically designed for simulating the aerodynamics of flapping insect flight. Leveraging a Fourier pseudo-spectral approach with volume penalization, FluSI aims to address the challenges of simulating complex, time-dependent geometries of insects with flapping wings in viscous flow environments. The framework is intended for deployment on high-performance computing systems with distributed memory architectures, providing scalability and parallel efficiency.

Methodology and Implementation

FluSI tackles the intractable problem of simulating the fluid dynamics around flapping wings by employing a volume penalization method integrated with the Fourier pseudo-spectral discretization of the incompressible Navier-Stokes equations. The geometrical complexity of the insect and its wings is handled through a mask function that captures both the shape and movement of the flapping body. The solution space incorporates a computational model for the dynamic geometry of the insect comprising modules that simulate an insect's flight mechanics and wing kinematics.

Several validation tests are conducted to assess FluSI's performance, including simulations of a sedimenting sphere, a flapping rectangular wing, hovering fruit fly, and a free-flight butterfly model. These validation efforts demonstrate FluSI's ability to reproduce established computational results with a high degree of accuracy and efficiency.

The software is tailored for execution on massively parallel architectures, utilizing the Message Passing Interface (MPI) for communication across distributed memory systems. Its implementation relies heavily on the P3DFFT library for performing three-dimensional Fast Fourier Transforms and ensures optimal utilization of computational resources.

Numerical and Computational Insights

FluSI demonstrates significant flexibility and precision through its capacity to simulate flow regimes up to moderate Reynolds numbers. The use of a pseudo-spectral discretization aids in the precise resolution of fluid dynamics without the need for artificial diffusion or dispersion—common in finite difference and volume methods. The numerical approach balances the modeling error from penalization with the discretization error—a concept critical for maintaining accuracy in simulating realistic aerodynamic scenarios.

Results and Implications

Detailed numerical results illustrate the efficacy of FluSI in simulating complex aerodynamic phenomena such as those exhibited by insects during flight. In multiple validation scenarios, the software robustly captures the dynamics of vortical structures and inertial forces prevalent in flapping flight.

Through its modular design, FluSI facilitates a wide range of applications, from basic fluid-structure interactions to advanced scenarios involving turbulent bluff-body interactions. This versatility opens possibilities for investigating biologically-inspired aerodynamics and flight mechanics in a controlled computational environment.

Future Directions

The paper hints at potential future enhancements, specifically around addressing the constraints imposed by equidistant Cartesian grids. This involves improving computational efficiency and flexibility across various insect geometries and flight conditions. Additionally, expanding the capability to handle higher Reynolds numbers might broaden the application spectrum of FluSI further into the realms of small unmanned aerial vehicles and other areas in bio-inspired engineering.

In summary, FluSI represents a sophisticated tool for simulating flapping flight in insects, with implications not just for biological studies, but for advancements in computational fluid dynamics and the development of bio-mimetic technologies.