Large Eddy Simulation of Gust Noise Sources in a Low Pressure Axial Compressor

TLDR: In this paper, a large eddy simulation (LES) method was used to predict the forces on the blades of a single-stage axial flow compressor without guide vanes.

Abstract: Axial compresso rs are often subject to poor inflow conditions which range from steady state , but spatially asymmetric velocity profiles due to imperfect intake geometry to time varying ingested vortices, turbulence or secondary inflow distortion s. The resulting periodic and/or random forces acting on the blades cause tonal and/or broadband interaction noise which sometimes dominates the noise radiation of the compressor . In th e present study, we appl y the large eddy simulation (LES) method to pre dict the unsteady forces on the blades of a single stage low pressure axial flow compressor without guide vanes. T urbulence is generated by a turbulence grid installed upstream of the impeller’s leading edge plane. The LES shows that the complex wake/vorte x flow downstream of the turbulence grid produces both , “random” (i.e. broadband) and periodic velocity fluctuations. The predicted blade response to the inflow turbulence (in terms of surface pressure fluctuations in the blade leading edge region) is in g ood agreement with our own surface pressure measurements. Assuming the impeller to be placed in an infinitely long duct a simple algebraic acoustic model proposed by Morfey is employed to predict the duct sound spectrum . By way of contrast, assuming a very short duct type inlet , the Ffowcs Williams and Hawkings analogy is employed to predict the spectral free field suction side sound pressure. Employing the LES data the acoustic calculation s yield a good first estimate of the absolute levels of the radiated sound. Also, the characteristic differences between clean inflow (i.e. without turbulence grid) and highly turbulent inflow are reasonably well predicted as compared to experiments. Discrepancies are attributed to insufficient boundary layer reso lution , t he omission of the tip clearance , and the limitations of the acoustical model s.