Docsity
Docsity

Prepare-se para as provas
Prepare-se para as provas

Estude fácil! Tem muito documento disponível na Docsity


Ganhe pontos para baixar
Ganhe pontos para baixar

Ganhe pontos ajudando outros esrudantes ou compre um plano Premium


Guias e Dicas
Guias e Dicas

Homogenous turbulence, Notas de estudo de Matemática

Study of homogenous isotropic turbulence in three-dimensions

Tipologia: Notas de estudo

2022

Compartilhado em 19/06/2023

luke-da-luz-sousa-9
luke-da-luz-sousa-9 🇧🇷

5 documentos

1 / 6

Toggle sidebar

Esta página não é visível na pré-visualização

Não perca as partes importantes!

bg1
An Internet Book on Fluid Dynamics
Stability of Planar Parallel Flows
In this section we examine some of the results obtained from the linear stability analyses described in
section (Bkc). It is convenient to begin with the data for boundary layers presented in Figures 1 to 4.
The equations governing the perturbations in section (Bkc) suggest that it is convenient and appropriate
to non-dimensionalize the coordinate xby using a surrogate parameter in the form of the displacement
thickness, δD, and to non-dimensionalize this in the form of δDU/ν which is a Reynolds number based
on the displacement thickness and the velocity, U, at the edge of the boundary layer. Thus δDU/ν will
represent the distance, x, in the data that follows. Moreover, a convenient non-dimensional frequency is
ωRν/U2.
Figure 1: Neutral stability boundary (kI= 0) for the Blasius boundary layer velocity profile in a graph of the perturbation
frequency, ωRν/U2, plotted against the surrogate distance parameter, δDU/ν.
As a first example of the results of a stability calculation, Figure 1 presents the neutral stability contour
(kI= 0) for the Blasius boundary layer velocity profile in a graph of the potential perturbation frequency,
ωRν/U2, plotted against the surrogate distance parameter, δDU/ν. The regions of stable behavior surround
a region of instability and the neutral stability line demonstrates that the boundary layer, which is stable
when δDis small, first becomes unstable when δDU/ν 600 as we can see by constructing a vertical
tangent to the neutral stability curve. Since we have seen in section (Bjd) that δD=1.72(νx/U)1
2for the
Blasius boundary layer it follows that this will first become unstable when the distance xfrom the leading
edge increases to a value given by
Ux
ν=(Rex)=600
1.722
=1.22 ×105(Bkd1)
where Rexdenotes the Reynolds number based on Uand x. It also follows that the first frequency which
becomes unstable is given by the point where the vertical tangent touches the neutral stability curve and
from Figure 1 that frequency is given by
ωR1.6×104U2
ν(Bkd2)
Since it is that frequency which is first amplified, it is usually that frequency that is observed experimentally
during the first stages of transition. This assumes that the inherent noise available for amplification is at
least broadband if not white.
pf3
pf4
pf5

Pré-visualização parcial do texto

Baixe Homogenous turbulence e outras Notas de estudo em PDF para Matemática, somente na Docsity!

An Internet Book on Fluid Dynamics

Stability of Planar Parallel Flows

In this section we examine some of the results obtained from the linear stability analyses described in section (Bkc). It is convenient to begin with the data for boundary layers presented in Figures 1 to 4. The equations governing the perturbations in section (Bkc) suggest that it is convenient and appropriate to non-dimensionalize the coordinate x by using a surrogate parameter in the form of the displacement thickness, δD, and to non-dimensionalize this in the form of δDU/ν which is a Reynolds number based on the displacement thickness and the velocity, U, at the edge of the boundary layer. Thus δDU/ν will represent the distance, x, in the data that follows. Moreover, a convenient non-dimensional frequency is ωRν/U^2.

Figure 1: Neutral stability boundary (kI = 0) for the Blasius boundary layer velocity profile in a graph of the perturbation frequency, ωRν/U 2 , plotted against the surrogate distance parameter, δD U/ν.

As a first example of the results of a stability calculation, Figure 1 presents the neutral stability contour (kI = 0) for the Blasius boundary layer velocity profile in a graph of the potential perturbation frequency, ωRν/U^2 , plotted against the surrogate distance parameter, δDU/ν. The regions of stable behavior surround a region of instability and the neutral stability line demonstrates that the boundary layer, which is stable when δD is small, first becomes unstable when δDU/ν ≈ 600 as we can see by constructing a vertical

tangent to the neutral stability curve. Since we have seen in section (Bjd) that δD = 1.72(νx/U)

1 (^2) for the Blasius boundary layer it follows that this will first become unstable when the distance x from the leading edge increases to a value given by ( Ux ν

= (Rex) =

= 1. 22 × 105 (Bkd1)

where Rex denotes the Reynolds number based on U and x. It also follows that the first frequency which becomes unstable is given by the point where the vertical tangent touches the neutral stability curve and from Figure 1 that frequency is given by

ωR ≈ 1. 6 × 10 −^4

U^2

ν

(Bkd2)

Since it is that frequency which is first amplified, it is usually that frequency that is observed experimentally during the first stages of transition. This assumes that the inherent noise available for amplification is at least broadband if not white.

Figure 2: As Figure 1, comparing the calculated neutral stability boundary with experimental measurements. Adapted from Schlichting (1960).

Figure 3: Spatial amplification for the Blasius boundary layer velocity profile in a graph of the perturbation frequency, ωRν/U 2 , plotted against the surrogate distance parameter, δD U/ν, showing contours of several amplification rates, −kI ν/U , as indicated. Adapted from Jaffe et al. (1970); see also Sherman (1990).

Figure 2 shows a comparison between the analytical neutral stability curve presented in Figure 1 and some experimental observations of that neutral stability boundary. The agreement is good and any discrepancy can probably be attributed to the difficulty of determining exactly where the instability begins.

To proceed beyond the neutral stability curve and calculate the growth of the noise downstream of the onset of amplification as we discuss in the next section (Bke), knowledge of the amplification rates inside the neutral stability curve are needed. For the Blasius boundary layer profile, some contours of non- zero non-dimensional amplification rates, −kI ν/U, are shown in Figure 3. How this data can be used is discussed in the section that follows.

Figure 5: Flow of water past an ogival headform showing the growing Tollmein-Schlichting waves on the interfacial layer just after separation (Brennen 1970).

Figure 6: Flow of water past a spherical headform showing the growing Tollmein-Schlichting waves on the interfacial layer just after separation (Brennen 1970).

maximum or centerline velocity of the flow.

Figure 7: Water jet emerging from a laminar nozzle (Hoyt and Taylor 1977).

Figure 8: General amplification chart for the spatial or temporal stability of an interfacial boundary layer with a free surface in a graph of the imaginary wavespeed, cI /ΔU , plotted against (U − cR )/ΔU. Contours of real and imaginery wavenumbers, bkR and bkI are shown. Adapted from Brennen (1970).

Figure 9: Planar Poiseuille flow between two flat plates.