Titolo della tesi: Development of a high-fidelity OpenFOAM solver for aircraft aerodynamics
The development of the rhoEnergyFoam solver by Modesti and Pirozzoli represents a significant advancement in the numerical simulation of compressible turbulent flows. The solver was designed to bridge the gap between high-accuracy academic methods and general-purpose Computational Fluid Dynamics (CFD) solvers commonly used in industry. By leveraging an energy-consistent, low-diffusive numerical flux and selectively incorporating diffusion based on flow characteristics, rhoEnergyFoam minimizes numerical dissipation while maintaining robustness. It utilizes a novel triple-splitting approach, which decomposes the advection fluxes of the Navier-Stokes equations via the AUSM (Advection Upstream Splitting Method) into distinct pressure and convective components. This formulation allows tailored numerical diffusion for various flow regimes: fully resolved simulations (Mode A) operating without artificial diffusion, unresolved smooth flows (Mode B) applying minimal pressure diffusion, and shocked flows (Mode C) using both pressure and convective diffusion to ensure stability. Implemented within the OpenFOAM framework, rhoEnergyFoam provides a robust, highly accurate tool for simulating compressible flows across a wide range of Mach and Reynolds numbers. The solver’s low-dissipation characteristics improve the accuracy of turbulence simulations, preserving kinetic energy in Unsteady RANS (URANS) and Detached Eddy Simulations (DES), and reducing numerical damping that can obscure physical flow features. Its shock-capturing capability, based on a local shock sensor, provides sharp resolution of supersonic flows, which is critical for aerospace and automotive applications. Validation through a variety of test cases has demonstrated that rhoEnergyFoam outperforms standard OpenFOAM solvers in preserving kinetic energy and in predicting aerodynamic coefficients such as lift and drag. By maintaining minimal numerical diffusion while selectively introducing it where necessary, the solver ensures high-fidelity flow predictions without sacrificing stability. Since its development, rhoEnergyFoam has been applied to a variety of complex flow problems. Li and Paoli employed the solver for simulations of ice accretion on aircraft wings, demonstrating its applicability in aerospace contexts. Zadeh and Paoli extended its use to RANS simulations of supersonic jets, confirming its performance in high-speed exhaust flows. Moreover, rhoEnergyFoam has been successfully validated against DNS/LES of compressible boundary layers and transitional flows, including hypersonic boundary layers at Mach 6. These studies underscore rhoEnergyFoam’s versatility and robustness in addressing aerospace-relevant fluid dynamics challenges. This industrial PhD project builds upon the foundation laid by rhoEnergyFoam by applying the solver to high-fidelity CFD simulations of aerospace flows. Specifically, the project focuses on integrating rhoEnergyFoam within a hybrid RANS/LES framework (e.g., DES) to simulate complex compressible flows in aerospace configurations, such as transonic and supersonic flows. The low-dissipation features of the solver are expected to provide more accurate predictions of shocked flows and transition phenomena in these applications. Through further validation and extension of rhoEnergyFoam’s capabilities, this project aims to contribute valuable insights to the aerospace CFD community and enhance industrial practices in the field.