Does the Kolmogorov scaling bridge hydrodynamic linear stability … · 2011. 11. 20. · Francesca...
Transcript of Does the Kolmogorov scaling bridge hydrodynamic linear stability … · 2011. 11. 20. · Francesca...
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Does the Kolmogorov scaling bridgehydrodynamic linear stability and turbulence?
Stefania ScarsoglioFrancesca De Santi Daniela Tordella
Department of Aeronautics and Space Engineering, Politecnico di Torino, Torino, Italy
American Physical Society
64th Annual DFD Meeting
November 20-22, 2011
Baltimore, Maryland
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Energy spectrum in fully developed turbulence
• Phenomenology of turbulence Kolmogorov 1941:−5/3 power-law for the energy spectrum over the inertial range;
• Common criterium for the production of a fully developed turbu-lent field to verify such a scaling (e.g. Frisch, 1995; Sreenivasan& Antonia, ARFM, 1997; Kraichnan, Phys. Fluids, 1967).
(left) Evangelinos & Karniadakis, JFM 1999. (right) Champagne, JFM 1978.
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Energy spectrum in fully developed turbulence
• Phenomenology of turbulence Kolmogorov 1941:−5/3 power-law for the energy spectrum over the inertial range;
• Common criterium for the production of a fully developed turbu-lent field to verify such a scaling (e.g. Frisch, 1995; Sreenivasan& Antonia, ARFM, 1997; Kraichnan, Phys. Fluids, 1967).
(left) Evangelinos & Karniadakis, JFM 1999. (right) Champagne, JFM 1978.
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Energy spectrum and linear stability analysis
• We study the state that precedes the onset of instability and tran-sition to turbulence:
• To understand how spectral representation can effectivelyhighlight the nonlinear interaction among different scales;
• To quantify the degree of generality on the value of the energydecay exponent of the inertial range;
• Different typical perturbed shear systems: plane Poiseuille flowand bluff-body wake.
• The set of small 3D perturbations:• Constitutes a system of multiple spatial and temporal scales;• Includes all the processes of the perturbative Navier-Stokes equa-
tions;• Leaves aside the nonlinear interaction among the different scales;
• The perturbative evolution is ruled by the initial-value problemassociated to the Navier-Stokes linearized formulation.
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Energy spectrum and linear stability analysis
• We study the state that precedes the onset of instability and tran-sition to turbulence:
• To understand how spectral representation can effectivelyhighlight the nonlinear interaction among different scales;
• To quantify the degree of generality on the value of the energydecay exponent of the inertial range;
• Different typical perturbed shear systems: plane Poiseuille flowand bluff-body wake.
• The set of small 3D perturbations:• Constitutes a system of multiple spatial and temporal scales;• Includes all the processes of the perturbative Navier-Stokes equa-
tions;• Leaves aside the nonlinear interaction among the different scales;
• The perturbative evolution is ruled by the initial-value problemassociated to the Navier-Stokes linearized formulation.
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Energy spectrum and linear stability analysis
• We study the state that precedes the onset of instability and tran-sition to turbulence:
• To understand how spectral representation can effectivelyhighlight the nonlinear interaction among different scales;
• To quantify the degree of generality on the value of the energydecay exponent of the inertial range;
• Different typical perturbed shear systems: plane Poiseuille flowand bluff-body wake.
• The set of small 3D perturbations:• Constitutes a system of multiple spatial and temporal scales;• Includes all the processes of the perturbative Navier-Stokes equa-
tions;• Leaves aside the nonlinear interaction among the different scales;
• The perturbative evolution is ruled by the initial-value problemassociated to the Navier-Stokes linearized formulation.
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Energy spectrum and linear stability analysis
• We study the state that precedes the onset of instability and tran-sition to turbulence:
• To understand how spectral representation can effectivelyhighlight the nonlinear interaction among different scales;
• To quantify the degree of generality on the value of the energydecay exponent of the inertial range;
• Different typical perturbed shear systems: plane Poiseuille flowand bluff-body wake.
• The set of small 3D perturbations:• Constitutes a system of multiple spatial and temporal scales;• Includes all the processes of the perturbative Navier-Stokes equa-
tions;• Leaves aside the nonlinear interaction among the different scales;
• The perturbative evolution is ruled by the initial-value problemassociated to the Navier-Stokes linearized formulation.
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Energy spectrum and linear stability analysis
• We study the state that precedes the onset of instability and tran-sition to turbulence:
• To understand how spectral representation can effectivelyhighlight the nonlinear interaction among different scales;
• To quantify the degree of generality on the value of the energydecay exponent of the inertial range;
• Different typical perturbed shear systems: plane Poiseuille flowand bluff-body wake.
• The set of small 3D perturbations:• Constitutes a system of multiple spatial and temporal scales;
• Includes all the processes of the perturbative Navier-Stokes equa-tions;
• Leaves aside the nonlinear interaction among the different scales;
• The perturbative evolution is ruled by the initial-value problemassociated to the Navier-Stokes linearized formulation.
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Energy spectrum and linear stability analysis
• We study the state that precedes the onset of instability and tran-sition to turbulence:
• To understand how spectral representation can effectivelyhighlight the nonlinear interaction among different scales;
• To quantify the degree of generality on the value of the energydecay exponent of the inertial range;
• Different typical perturbed shear systems: plane Poiseuille flowand bluff-body wake.
• The set of small 3D perturbations:• Constitutes a system of multiple spatial and temporal scales;• Includes all the processes of the perturbative Navier-Stokes equa-
tions;
• Leaves aside the nonlinear interaction among the different scales;
• The perturbative evolution is ruled by the initial-value problemassociated to the Navier-Stokes linearized formulation.
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Energy spectrum and linear stability analysis
• We study the state that precedes the onset of instability and tran-sition to turbulence:
• To understand how spectral representation can effectivelyhighlight the nonlinear interaction among different scales;
• To quantify the degree of generality on the value of the energydecay exponent of the inertial range;
• Different typical perturbed shear systems: plane Poiseuille flowand bluff-body wake.
• The set of small 3D perturbations:• Constitutes a system of multiple spatial and temporal scales;• Includes all the processes of the perturbative Navier-Stokes equa-
tions;• Leaves aside the nonlinear interaction among the different scales;
• The perturbative evolution is ruled by the initial-value problemassociated to the Navier-Stokes linearized formulation.
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Energy spectrum and linear stability analysis
• We study the state that precedes the onset of instability and tran-sition to turbulence:
• To understand how spectral representation can effectivelyhighlight the nonlinear interaction among different scales;
• To quantify the degree of generality on the value of the energydecay exponent of the inertial range;
• Different typical perturbed shear systems: plane Poiseuille flowand bluff-body wake.
• The set of small 3D perturbations:• Constitutes a system of multiple spatial and temporal scales;• Includes all the processes of the perturbative Navier-Stokes equa-
tions;• Leaves aside the nonlinear interaction among the different scales;
• The perturbative evolution is ruled by the initial-value problemassociated to the Navier-Stokes linearized formulation.
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Spectral analysis through initial-value problem
• The linear transient dynamics offers a great variety of very differ-ent behaviors (Scarsoglio et al., 2009, 2010, 2011):
⇒ Understand how the energy spectrum behaves;
• Is the linearized perturbative system able to show a power-law scaling for the energy spectrum in an analogous way tothe Kolmogorov argument?
• We determine the energy decay exponent of arbitrary pertur-bations in their asymptotic states and we compare it with the-5/3 Kolmogorov decay.
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Spectral analysis through initial-value problem
• The linear transient dynamics offers a great variety of very differ-ent behaviors (Scarsoglio et al., 2009, 2010, 2011):⇒ Understand how the energy spectrum behaves;
• Is the linearized perturbative system able to show a power-law scaling for the energy spectrum in an analogous way tothe Kolmogorov argument?
• We determine the energy decay exponent of arbitrary pertur-bations in their asymptotic states and we compare it with the-5/3 Kolmogorov decay.
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Spectral analysis through initial-value problem
• The linear transient dynamics offers a great variety of very differ-ent behaviors (Scarsoglio et al., 2009, 2010, 2011):⇒ Understand how the energy spectrum behaves;
• Is the linearized perturbative system able to show a power-law scaling for the energy spectrum in an analogous way tothe Kolmogorov argument?
• We determine the energy decay exponent of arbitrary pertur-bations in their asymptotic states and we compare it with the-5/3 Kolmogorov decay.
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Spectral analysis through initial-value problem
• The linear transient dynamics offers a great variety of very differ-ent behaviors (Scarsoglio et al., 2009, 2010, 2011):⇒ Understand how the energy spectrum behaves;
• Is the linearized perturbative system able to show a power-law scaling for the energy spectrum in an analogous way tothe Kolmogorov argument?
• We determine the energy decay exponent of arbitrary pertur-bations in their asymptotic states and we compare it with the-5/3 Kolmogorov decay.
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Perturbation scheme
• Linear 3D perturbative equations in terms of velocity and vorticity(Criminale & Drazin, Stud. Appl. Math., 1990);
• Laplace-Fourier (wake) and Fourier-Fourier (channel) transformin the x and z directions.
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Perturbation scheme
• Linear 3D perturbative equations in terms of velocity and vorticity(Criminale & Drazin, Stud. Appl. Math., 1990);
• Laplace-Fourier (wake) and Fourier-Fourier (channel) transformin the x and z directions.
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Perturbation scheme
• Linear 3D perturbative equations in terms of velocity and vorticity(Criminale & Drazin, Stud. Appl. Math., 1990);
• Laplace-Fourier (wake) and Fourier-Fourier (channel) transformin the x and z directions.
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Perturbative equations
• Perturbative linearized system:
∂2v
∂y2− k2v = Γ
∂Γ
∂t= iα
(d2U
dy2v − U Γ
)+
1
Re
(∂2Γ
∂y2− k2Γ
)∂ωy
∂t= −iαUωy − iγ
dU
dyv +
1
Re
(∂2ωy
∂y2− k2ωy
)The transversal velocity and vorticity components are v and ωy
respectively, Γ is defined as Γ = ∂xωz − ∂zωx.
• Initial conditions: symmetric and asymmetric inputs;
• Boundary conditions: (u, v, w)→ 0 as y → ±∞ and at walls.
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Perturbative equations
• Perturbative linearized system:
∂2v
∂y2− k2v = Γ
∂Γ
∂t= iα
(d2U
dy2v − U Γ
)+
1
Re
(∂2Γ
∂y2− k2Γ
)∂ωy
∂t= −iαUωy − iγ
dU
dyv +
1
Re
(∂2ωy
∂y2− k2ωy
)The transversal velocity and vorticity components are v and ωy
respectively, Γ is defined as Γ = ∂xωz − ∂zωx.
• Initial conditions: symmetric and asymmetric inputs;
• Boundary conditions: (u, v, w)→ 0 as y → ±∞ and at walls.
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Perturbative equations
• Perturbative linearized system:
∂2v
∂y2− k2v = Γ
∂Γ
∂t= iα
(d2U
dy2v − U Γ
)+
1
Re
(∂2Γ
∂y2− k2Γ
)∂ωy
∂t= −iαUωy − iγ
dU
dyv +
1
Re
(∂2ωy
∂y2− k2ωy
)The transversal velocity and vorticity components are v and ωy
respectively, Γ is defined as Γ = ∂xωz − ∂zωx.
• Initial conditions: symmetric and asymmetric inputs;
• Boundary conditions: (u, v, w)→ 0 as y → ±∞ and at walls.
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Perturbative equations
−1 −0.5 0 0.5 1
0
0.2
0.4
0.6
0.8
1
y
U(y)asym inputsym input
Channelflow
−5 0 50.4
0.6
0.8
1
y
sym input asym input
Wake flow
U(y)Re=30,x
0=10
U(y)Re=30,x
0=50
U(y)Re=100,x
0=10
U(y)Re=100,x
0=50
• Initial conditions: symmetric and asymmetric inputs;
• Boundary conditions: (u, v, w)→ 0 as y → ±∞ and at walls.
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Perturbation energy
• Kinetic energy density e:
e(t;α, γ) =1
2
∫ +yd
−yd
(|u|2 + |v|2 + |w|2)dy
• Amplification factor G:
G(t;α, γ) =e(t;α, γ)
e(t = 0;α, γ)
• Temporal growth rate r:
r(t;α, γ) =log[e(t;α, γ)]
2t
• Angular frequency (pulsation) ω:
ω(t; y = y0, α, γ) =dϕ(t; y = y0, α, γ)
dt, ϕ time phase
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Perturbation energy
• Kinetic energy density e:
e(t;α, γ) =1
2
∫ +yd
−yd
(|u|2 + |v|2 + |w|2)dy
• Amplification factor G:
G(t;α, γ) =e(t;α, γ)
e(t = 0;α, γ)
• Temporal growth rate r:
r(t;α, γ) =log[e(t;α, γ)]
2t
• Angular frequency (pulsation) ω:
ω(t; y = y0, α, γ) =dϕ(t; y = y0, α, γ)
dt, ϕ time phase
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Perturbation energy
• Kinetic energy density e:
e(t;α, γ) =1
2
∫ +yd
−yd
(|u|2 + |v|2 + |w|2)dy
• Amplification factor G:
G(t;α, γ) =e(t;α, γ)
e(t = 0;α, γ)
• Temporal growth rate r:
r(t;α, γ) =log[e(t;α, γ)]
2t
• Angular frequency (pulsation) ω:
ω(t; y = y0, α, γ) =dϕ(t; y = y0, α, γ)
dt, ϕ time phase
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Perturbation energy
• Kinetic energy density e:
e(t;α, γ) =1
2
∫ +yd
−yd
(|u|2 + |v|2 + |w|2)dy
• Amplification factor G:
G(t;α, γ) =e(t;α, γ)
e(t = 0;α, γ)
• Temporal growth rate r:
r(t;α, γ) =log[e(t;α, γ)]
2t
• Angular frequency (pulsation) ω:
ω(t; y = y0, α, γ) =dϕ(t; y = y0, α, γ)
dt, ϕ time phase
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Relevant transient behaviors
0 1000 3000 40000
t
G
2x104
104
k=2k=1.5
k=1Channel flow
Re=10000Sym ic, φ=π/2
2.5<k<1000
0 500 1000 1500
10−10
10−5
100
t
G
Wake flow Asym ic φ=0 Re=100 x
0= 50
0.45≤ k≤ 500
0 10 20 300
5
10
15
t
ω
Sym φ=0Asym φ=0Sym φ=π/4Asym φ=π/4
Channel flow k=15Re=10000 y
0=0.5
0 500 1000 1500
0.3
0.6
ω
t
Sym φ=0Asym φ=0Sym φ=π/4Asym φ=π/4
Wake flowRe= 100x
0=50 k=0.7
y0=1
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Spectral representation
• The energy spectrum is evaluated as the wavenumber distributionof the amplification factor, G(k);
• The spectral representation is determined by comparing theenergy of the waves when they are exiting their transientstate;
• Every perturbation has a characteristic transient exit time, Te;
• The asymptotic condition is reached when the perturbative waveexceeds the transient exit time, Te, that is when r ∼ const issatisfied for stable and unstable waves.
Scarsoglio, De Santi & Tordella, submitted to Phys. Rev. Lett., 2011.
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Spectral representation
• The energy spectrum is evaluated as the wavenumber distributionof the amplification factor, G(k);
• The spectral representation is determined by comparing theenergy of the waves when they are exiting their transientstate;
• Every perturbation has a characteristic transient exit time, Te;
• The asymptotic condition is reached when the perturbative waveexceeds the transient exit time, Te, that is when r ∼ const issatisfied for stable and unstable waves.
Scarsoglio, De Santi & Tordella, submitted to Phys. Rev. Lett., 2011.
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Spectral representation
• The energy spectrum is evaluated as the wavenumber distributionof the amplification factor, G(k);
• The spectral representation is determined by comparing theenergy of the waves when they are exiting their transientstate;
• Every perturbation has a characteristic transient exit time, Te;
• The asymptotic condition is reached when the perturbative waveexceeds the transient exit time, Te, that is when r ∼ const issatisfied for stable and unstable waves.
Scarsoglio, De Santi & Tordella, submitted to Phys. Rev. Lett., 2011.
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Spectral representation
• The energy spectrum is evaluated as the wavenumber distributionof the amplification factor, G(k);
• The spectral representation is determined by comparing theenergy of the waves when they are exiting their transientstate;
• Every perturbation has a characteristic transient exit time, Te;
• The asymptotic condition is reached when the perturbative waveexceeds the transient exit time, Te, that is when r ∼ const issatisfied for stable and unstable waves.
Scarsoglio, De Santi & Tordella, submitted to Phys. Rev. Lett., 2011.
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Energy G(k) at the asymptotic state (r ∼const)
100
101
102
103
10−4
10−8
100
103
k
G
k−5/3 k−5/3
k−2
MKM
Channel flowRe=500
100
101
102
103
10−5
100
105
k
G
SV
Channel flowRe=10000
k−5/3k−2
k−5/3
Present results (Re values: see panels)Asym ic, φ=π/2 Asym ic, φ=π/4 Asym, φ=0 Sym ic, φ=π/2 Sym ic, φ=π/4 Sym, φ=0
(Ex, E
y, E
z) Moser et al. (MKM), 1999, Re=12390, DNS
Saddoughi & Veeravalli (SV), 1993, Reδ*=4994, lab data
(Eu, E
v) Ong & Wallace (OW), 1996, Re=3900, lab data
Cerutti & Meneveau (CM), 2000, Re=572080, DNS & LES
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Energy G(k) at the asymptotic state (r ∼const)
100
101
102
103
10−8
10−4
100
104
k
G
k−5/3
k−5/3
CM
k−2Wake flowRe=30 x
0=50
100
101
102
103
10−5
100
105
k
G
OW
k−5/3
Wake flowRe=100 x
0=10
k−5/3
k−2
Present results (Re values: see panels)Asym ic, φ=π/2 Asym ic, φ=π/4 Asym, φ=0 Sym ic, φ=π/2 Sym ic, φ=π/4 Sym, φ=0
(Ex, E
y, E
z) Moser et al. (MKM), 1999, Re=12390, DNS
Saddoughi & Veeravalli (SV), 1993, Reδ*=4994, lab data
(Eu, E
v) Ong & Wallace (OW), 1996, Re=3900, lab data
Cerutti & Meneveau (CM), 2000, Re=572080, DNS & LES
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Transient exit time Te(k)
100
101
102
103
10−2
100
102
k
Te
k−2
k−5/3
Channel flowRe=500
100
101
102
10310
−2
100
102
104
k
Te
k−5/3
k−2Channel flowRe=10000
100
101
102
103
10−2
100
102
k
Te
k−5/3
k−2
Wake flow Re=30 x
0=50
100
101
102
10310
−3
100
103
Te
k
Wake flowRe=100 x
0=10
k−2
k−5/3
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Concluding remarks
• Spectrum determined by evaluating the energy of the waves whenthey are exiting their transient state;
• Regardless the symmetry and obliquity of perturbations, there ex-ists an intermediate range of wavenumbers in the spectrumwhere the energy decays with the same exponent observedfor fully developed turbulent flows (−5/3), where the nonlinearinteraction is considered dominant;
• Scale-invariance of G and Te at different (stable and unstable)Reynolds numbers and for different shear flows;
• The −5/3 spectral power-law scaling of inertial waves seemsto be a general and intrinsic dynamical property of the NSsolutions encompassing the nonlinear interaction.
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Concluding remarks
• Spectrum determined by evaluating the energy of the waves whenthey are exiting their transient state;
• Regardless the symmetry and obliquity of perturbations, there ex-ists an intermediate range of wavenumbers in the spectrumwhere the energy decays with the same exponent observedfor fully developed turbulent flows (−5/3), where the nonlinearinteraction is considered dominant;
• Scale-invariance of G and Te at different (stable and unstable)Reynolds numbers and for different shear flows;
• The −5/3 spectral power-law scaling of inertial waves seemsto be a general and intrinsic dynamical property of the NSsolutions encompassing the nonlinear interaction.
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Concluding remarks
• Spectrum determined by evaluating the energy of the waves whenthey are exiting their transient state;
• Regardless the symmetry and obliquity of perturbations, there ex-ists an intermediate range of wavenumbers in the spectrumwhere the energy decays with the same exponent observedfor fully developed turbulent flows (−5/3), where the nonlinearinteraction is considered dominant;
• Scale-invariance of G and Te at different (stable and unstable)Reynolds numbers and for different shear flows;
• The −5/3 spectral power-law scaling of inertial waves seemsto be a general and intrinsic dynamical property of the NSsolutions encompassing the nonlinear interaction.
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Concluding remarks
• Spectrum determined by evaluating the energy of the waves whenthey are exiting their transient state;
• Regardless the symmetry and obliquity of perturbations, there ex-ists an intermediate range of wavenumbers in the spectrumwhere the energy decays with the same exponent observedfor fully developed turbulent flows (−5/3), where the nonlinearinteraction is considered dominant;
• Scale-invariance of G and Te at different (stable and unstable)Reynolds numbers and for different shear flows;
• The −5/3 spectral power-law scaling of inertial waves seemsto be a general and intrinsic dynamical property of the NSsolutions encompassing the nonlinear interaction.
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Coming next...
• Analysis of the perturbation transient dynamics in the 2Dand 3D boundary layer (W. O. Criminale, University of Wash-ington);
• Analytical integration of the kinetic energy equation basedon the perturbed velocity and vorticity field (G. Staffilani,MIT)⇒ Study of the intermediate and long term asymptotics.
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?
Does theKolmogorov
scaling bridgehydrodynamiclinear stability
and turbulence?
S. Scarsoglio,F. De Santi,D. Tordella
Introduction
Motivation andgeneral aspects
Initial-valueproblem
Mathematicalframework
Measure of thegrowth
Linear transientdynamics
Results
Perturbativesystem features
Spectraldistributions
Conclusions
Conclusions
Coming next...
• Analysis of the perturbation transient dynamics in the 2Dand 3D boundary layer (W. O. Criminale, University of Wash-ington);
• Analytical integration of the kinetic energy equation basedon the perturbed velocity and vorticity field (G. Staffilani,MIT)⇒ Study of the intermediate and long term asymptotics.
S. Scarsoglio, F. De Santi, D. Tordella Does the Kolmogorov scaling bridge hydrodynamic linear stability and turbulence?