INDUSTRIAL CFD SIMULATION OF AERODYNAMIC NOISE

Abstract

Real challenges to suppress undesirable fluid-excited acoustics are posed by a wide variety of engineering disciplines. Noise regulations, passenger comfort and component stability are motivators which are continuing to stimulate substantial efforts towards the understanding of aeroacoustic phenomena, and not least to quantify the usability (practicability and value) of traditional and advanced prediction methods. The latter is the primary focus of this thesis, particularly as applied to the transportation industries, aerospace, automotive and rail. Nowadays Computational Fluid Dynamics (CFD) is a tool well integrated into the industrial development and production life-cycles. This is possible now because of two main factors: the increase in the performance of relatively cheap personal computers and network facilities, and the progress made in general purpose CFD software between modeling complexity and practicability within the industrial environment. While CFD methodologies are well established for lots of applications such as aerodynamics, heat exchange, etc., aeroacoustic CFD simulations still represent a challenge, in particular their industrial practicability. In these years this has given rise to heavy investments by the automotive industry in international aeroacoustics consortia, whereby all the major car companies work together to study the limitations and advantages of aeroacoustics CFD. The general aim of these consortia is to develop methodologies which fit into, and improve upon, current design processes. The goal of the present work is to explore the multitude of different CFD modeling approaches for some typical industrial problems such as: cavity noise, vortex shedding noise, propeller and jet noise. Each of these problems has a particular mechanism for noise generation and different methods have been studied and tested, in order to develop and optimize a practical methodology for the analysis of each problem type. Furthermore each of the aeroacoustics problems considered are representative of a variety of industrial applications. Cavity noise is at the origin of phenomena such as sun-roof buffeting in convertibles or door-gap tonal noise. Vortex shedding noise is typical of any flows involving bluff bodies such as automobile antennas or aircraft landing gear. Propeller noise is typical to applications involving rotating machinery, such as fans, pumps and turbines. Different approaches ranging from steady and transient RANS simulations with the acoustic analogy (including porous and solid surface formulations), to Computational Aero Acoustics (CAA) and Large Eddy Simulation (LES) type computations have been studied and applied. Classic theories already exist to predict aerodynamically generated noise, which are both computationally and economically less expensive than CFD methods. However aeroacoustics CFD is the future, beginning as a promising present, for the following reasons:  Industries are interested in modeling complex geometries.  Many classic theories can be applied successfully but very often restrictions exist with respect to the configuration and flow conditions. For example, classic propeller theories cannot be used to model real-world configurations such as a propeller installed on a wing with some prescribed yaw or angle of attack.  The progress of all other Computer Aided Design and Engineering tools, such as linear or non-linear structural codes, are driving design towards a virtual multi-physics approach for the simulation of complex geometries. Due to previous experience and the wide availability of modeling options, it was decided to use the general purpose CFD software package ANSYS FLUENT for CFD investigations in this study.