Numerical Modelling of Aerodynamic Noise in Compressible Flows

The solution of the AeroAcoustics (CAA) problems by means of the Direct Numerical Simulation (DNS) or even the Large Eddy Simulation (LES) for a large computational domain is very time consuming and cannot be applied widely for engineering applications. In this paper the in-house CFD and CAA codes are presented. The in-house CFD code is based on the LES approach whereas the CAA code is an acoustic postprocessor solving the non-linearized Euler equations for fluctuating (acoustic) variables. These codes are used to solve the aerodynamically generated sound field by a flow over a rectangular cavity with inlet Mach number 0.53.

• Computational Aeroacoustics; Noise
• Pressure Waves; Cavity Noise

1. Introduction

The compressible Navier-Stokes equations can numerically predict the aerodynamic as well as acoustic flow field simultaneously. The solution of the Navier-Stokes equations using DNS or LES methods for capturing both the aerodynamic and acoustic fluctuations is still very time consuming. Since there is a large disparity of the length and time scales between the aerodynamic and acoustic variables, the DES and LES methods are usually used for the source domain, it means there, were the aerodynamic disturbances generate noise. For the rest of computational domain, where the acoustic waves propagate the other methods can be used, e.g. the non-linearized Euler equations for fluctuating (acoustic) variables [1,2]. Usually, in the propagation region it is assumed that the flow field does not generate any sound. However, the form of the non-linearized Euler equations applied in the in-house acoustic postprocessor allows for such possibility. The non-linearized Euler equations can be applied for wide range of means flow Mach number and for internal flows. For the computational domain, called often as an acoustic source region, where flow disturbances generate noise, for better modeling acoustic excitations, the LES method was implemented into the compressible in-house CFD code. This CFD code is dedicated for modeling the flow with mean flow Mach number higher than 0.15.

The in-house CFD/CAA codes have been used for computations of the cavity flow noise applications, where both broadband and tonal noise is emitted. This type of flow is commonly found in aviation, turbomachinery, power engineering and of course in environmental flows. In the presented results, the aerodynamic and acoustic fields were considered as 2D. This assumption was taken in accordance with experiment [3] and also in order to decrease the computational domain. This paper focuses mainly on the qualitative assessment of the applied techniques and description of the appeared problems.

The phenomenon of flow induced noise radiation in cavities has been studied experimentally [3,4] by many researchers since many years. The validation of the inhouse acoustic postprocessor against Prof. Weyna experiments was preliminary carried out [2].

Figure 1 shows a schematic overview of the phenomena undergoing during the cavity flow with mean flow Mach number lower than 1. The pressure wave travelling inside the cavity induces the shear layer oscillation, what is mainly responsible for the acoustic wave propagation. In Figure 1 the cavity depth and length used for calculations was presented also (according to experiment of Ahuja and Mendoza [3].

References

1. S. Dykas, W. Wróblewski and T. Chmielniak, “Using a CFD/CAA Technique for Aerodynamic Noise Assessment,” Proceedings of ASME Turbo Expo 2008: Power for Land, Sea and Air GT2008, Berlin, 9-13 June 2008, pp. 784-794.

2. S. Dykas, W. Wróblewski, S. Rulik and T. Chmielniak, “Numerical Modeling of the Acoustic Waves Propagation,” Archives of Acoustics, Vol. 35, No. 1, 2010, pp. 35-48.

3. K. K. Ahuja and J. Mendoza, “Effects of Cavity Dimensions, Boundary Layer, and Temperature on Cavity Noise with Emphasis on Benchmark Data to Validate Computational Aeroacoustic Codes,” 1964.

4. S. Weyna, “An Acoustic Energy Dissipation of the Real Sources (in Polish),” 2005.

5. W. Wróblewski, S. Dykas and A. Gepert, “Modeling of the Flows with Condensation (in Polish),” 2006.