C OMPRESSIBILITY EFFECTS ON SOUND SOURCE DISTRIBUTIONS IN ISOTROPIC COMPRESSIBLE TURBULENCE
C OMPRESSIBILITY EFFECTS ON SOUND SOURCE DISTRIBUTIONS IN ISOTROPIC COMPRESSIBLE TURBULENCE
复制标题
DOI：
--
复制
发表时间：
2014
期刊：
影响因子：
--
通讯作者：
D. Terakado;T. Nonomura;M. Sato;K. Fujii
中科院分区：
文献类型：
--
作者：
D. Terakado;T. Nonomura;M. Sato;K. Fujii
Two turbulent mixing noises exist in supersonic jets. One is the Mach waves which are generated from large scale structures, and the other is the acoustic waves from fine scale structures. The two turbulent mixing noises are important, because they appear in ideally expanded jets, under expanded jets and over expanded jets (Tamet al., 1996). Many researchers have devoted in studying Mach waves. Williams (1963) showed that the far field acoustic power is proportional to the third power of jet velocity. Tam and Burton (1984a,b) developed the model for the mechanism of Mach wave generation in which wall like instability waves moving supersonically are assumed as the sound sources. Lately, the model was validated by comparing with experiments (Tam and Chen, 1992). Recently, in the frame of computational aeroacoustics (CAA) large-eddy simulation (LES) has been one of the strongest tools to predict Mach waves, because Mach waves are generated from large scale structures of turbulence. The prediction accuracy of over all sound pressure levels (OSPL) by LES are within 2dB for moderate Mach number cases (MJ ∼ 1.5) (Bodony and Lele, 2005), and within 5dB for high Mach number cases (MJ ∼ 4.0) (Nonomura et al., 2014). Note that the errors contain the contributions from acoustic waves from fine scale structures so that the prediction accuracy for Mach waves should be higher than values above. Acoustic waves from fine scale structures, on the other hand, has been much less understood than those of Mach waves. Only limited knowledge was given by the past studies. Tamet al. (1996) organized a huge amount of experimental data and identified the shape of spectra of acoustic waves from fine scale structures. They showed that the spectra spread wide range of frequency compared with the spectra of Mach waves which have a specific peak. Seror et al. (2000, 2001) investigated the contributions from sub-grid scale to the spectra of SPL. They adopted the idea of sub-grid scale model to the Lighthill’s acoustic analogy for the far field projection coupled with LES, and compared with the results of the filtered direct numerical simulations (DNS) with full-scale Lighthill’s analogy. The model proposed by Seror et al. recoverd the acoustic intensity lost in the filtering procedure and improved the prediction accuracy. Although those studies gave us some useful knowledge to understand the acoustic waves from fine scale structures, the detail mechanisms remain unclarified.
