Turbulence and sound-field POD analysis of a turbulent jet
Turbulence and sound-field POD analysis of a turbulent jet
复制标题
DOI：
10.1260/147547209787548903
复制
发表时间：
2009-01-01
影响因子：
1
通讯作者：
Colonius, T.
中科院分区：
文献类型：
--
作者：
Freund, J. B.;Colonius, T.
利用经过验证的直接数值模拟数据库，建立了马赫数为0.9的湍流射流的本征正交分解(POD)。范数的定义基于压力、湍动能、流向速度和总焓的近场体积积分，流向速度的二维积分(与实验测量相匹配)，以及球面上的远场压力积分。我们发现对于不同的范数，POD模式有很大的不同，它们在表示完整数据时的效率强烈地依赖于范数，特别是我们试图表示的数据。要再现近场湍流统计数据，需要由动能或压力范数计算的相对较少的模式。然而，需要使用近场范数计算的大量POD模式来表示声场。用近场声压范数或声场范数计算的主要近场POD模都具有波包结构。
A Proper Orthogonal Decomposition (POD) is constructed for a Mach 0.9 turbulent jet using a well-validated direct numerical simulation database. Norms are defined based on near-field volume integrals of pressure, turbulence kinetic energy, streamwise velocity, and total enthalpy, two-dimensional integrals of streamswise velocity (to match experimental measurements), and far-field integrals of pressure over a sphere. We find substantially different POD modes for the different norms, and their efficiency at representing the full data is strongly dependent upon the norm and specifically which data we attempt to represent. To reproduce near-field turbulence statistics requires relatively few modes computed by a kinetic energy or pressure norm. However, a large number of the POD modes computed using a near-field norm are required to represent the sound field. The dominant near-field POD modes computed with either the near-field pressure norm or the sound field norm have the structure of wave packets.