Pith. sign in

REVIEW 2 major objections 6 minor 121 references

Stochastic dynamical modeling of turbulent flows

T0 review · 2 major / 6 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This review argues that stochastically forced linearized Navier-Stokes equations, with colored-in-time forcing chosen by a convex covariance-completion problem, reproduce the second-order statistics of turbulent channel flow.

desk verdict A clear, honest invited review of the authors' own covariance-completion framework, with no new results and one genuine soft spot: the temporal spectrum is underdetermined by the covariance data. read the letter →

arxiv 1908.09487 v1 pith:3IV3DTWO submitted 2019-08-26 physics.flu-dyn cs.LGcs.SYeess.SYmath.OC

Armin Zare , Tryphon T. Georgiou , Mihailo R. Jovanović This is my paper · ORCID
classification physics.flu-dyncs.LGcs.SYeess.SYmath.OC
keywords turbulentflowsstochasticdynamicslinearizedNavier-Stokesequationscolored-in-timeforcingcovariancecompletionconvexoptimizationcontrol-orientedmodelingchannelflow
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This review argues that a linearized model of turbulent channel flow, driven by suitably colored-in-time stochastic forcing, can reproduce the second-order statistics that direct numerical simulation produces. The key idea is to treat the physics missing from linearization, mainly the nonlinear interactions, as an unknown input spectrum and to recover that spectrum from data by solving a convex optimization problem. The authors show that white-in-time forcing cannot work, because it forces a sign-definite covariance equation while turbulence requires sign-indefinite excitation. If the framework holds, turbulence models suited for control and estimation can be built from first-principles linearization plus limited statistical data, rather than from purely empirical closures.

What carries the argument

The load-bearing object is the covariance completion problem CC-1, which minimizes -log det(X) + gamma times the nuclear norm of Z subject to AX + XA* + Z = 0 and to the constraint that the available entries of CXC* equal the measured velocity covariances. Here A is the linearized Navier-Stokes generator around the turbulent mean velocity, X is the steady-state covariance of fluctuations, and Z = BH* + HB* is the generally sign-indefinite contribution of stochastic excitation. The log-det term enforces positive definiteness and maximum entropy, while the nuclear norm promotes a low-rank Z, which keeps the forcing model low-complexity and avoids trivial full-rank cancellations of the dynamics.

What would settle it

Compute the full two-point space-time correlation or spectral density S_vv(k,omega) from DNS for the same channel at Re=186 and wavenumber k=(2.5,7), and compare it with the model PSD from Equation 22; a mismatch in bandwidth, spectral shape, or lagged covariance R_vv(k,tau) for nonzero tau would show that the reported spatio-temporal spectrum is not determined by the matched covariances.

Watch

Extended reading notes

Core claim

The central claim is that all available one-point velocity correlations of turbulent channel flow can be reproduced by a stochastically forced linearized Navier-Stokes model, and that a suitably regularized solution also recovers a large share of the full covariance structure. At friction Reynolds number 186, solving the convex problem CC-1 matches every one-point correlation used as data at every wavenumber, and with regularization parameter gamma equal to 300 it recovers about 60 percent of the DNS covariance matrix. The same construction yields a spatio-temporal energy spectrum concentrated near y+ approximately 15, in line with DNS trends, and a PSD peak closer to DNS than the plain linearized model or an eddy-viscosity-enhanced linearized model.

Load-bearing premise

The load-bearing premise is that the particular colored-in-time filter chosen after the covariance fit, built from spatially and temporally uncorrelated white noise and one factorization of the excitation matrix, correctly captures temporal correlations; the covariance data alone do not fix this choice.

Editorial extensions

If this is right

  • White-in-time stochastic forcing cannot explain turbulent second-order statistics in wall-bounded flows, so turbulence models built on linearization must include colored-in-time excitation.
  • With low-rank regularization, only a small number of colored-in-time inputs are needed: at the most energetic wavenumber and gamma equal to 10^4, six inputs suffice, giving a tractable low-dimensional stochastic model.
  • The colored-in-time forcing is equivalent to a rank-limited state-feedback perturbation of the linearized generator, offering a dynamical correction that stands in for the omitted nonlinear interactions.
  • The recovered spatio-temporal spectrum reproduces DNS trends in inner units, suggesting the approach can produce physically meaningful temporal correlations, not just steady-state covariances.
  • A minimum-energy variant, CC-2, casts the same completion as an optimal state-feedback synthesis problem, which is directly compatible with control-oriented modeling.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The paper does not establish uniqueness of the spatio-temporal spectrum, because the covariance data fix X and Z but not the factorization Z = BH* + HB* or the white-noise covariance Omega; a future version could add spectral-density or lagged-covariance constraints to remove this freedom.
  • The same convex completion framework should transfer to other translationally invariant flows such as Couette flow, boundary layers, or jets, whenever mean-velocity linearization and partial second-order statistics are available; this is a direct testable extension.
  • The identified low-rank feedback correction could be compared against resolvent modes or spectral proper orthogonal decomposition modes to see whether the data-driven correction is capturing the same coherent structures that appear in DNS.
  • If the correction remains low-rank and robust across Reynolds numbers, it could serve as a reduced-order model for real-time estimation and drag-reduction control, though the paper does not yet demonstrate closed-loop performance.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

2 major / 6 minor

Summary. This review article presents a framework for constructing low-complexity stochastic models of turbulent channel flow whose second-order statistics match partially available DNS data. The dynamics are the Navier-Stokes equations linearized around a turbulent mean, and the missing effects of nonlinearity are represented by colored-in-time stochastic forcing generated by a linear filter. The central mathematical objects are two convex optimization problems, CC-1 and CC-2, which enforce Lyapunov-like consistency and data constraints while promoting low-rank or sparse forcing models. A channel flow case study at Re=186 demonstrates exact reproduction of one-point correlations, partial recovery of off-diagonal two-point covariances, stochastic linear simulations, and a spatio-temporal power spectral density comparison against DNS.

Significance. If the claims hold, the framework provides a useful bridge between data-driven and physics-based modeling by converting covariance completion into a convex optimization problem and by giving an efficient reduced-order simulation model through the minimal realization in Equation (22). The mathematical scaffold, including the admissibility conditions, the Lyapunov-like constraints, and the convexity statements, is presented clearly and is grounded in prior peer-reviewed work. The paper is also honest about several limitations, notably in Section 4.4 and in the Future Issues list. However, the spatio-temporal spectral comparison is currently conditional on an unspecified realization choice among infinitely many equivalent covariance completions, and the quantitative two-point recovery is moderate; these issues weaken the empirical case even though the theoretical framework remains sound.

major comments (2)
  1. [§4.2.1, Eq. (19); §5.3, Fig. 11(a)] The spatio-temporal spectrum reported in Figure 11(a) is not determined by the covariance-fitting data. The solution of CC-1 fixes X and Z, but Equation (17b) only fixes Z = B H* + H B*; for every factorization (B,H) and every admissible white-noise covariance Ω, the gain K in Equation (19) reproduces the fitted X exactly because the Lyapunov balance reduces to A X + X A* + Z = 0. The output spectrum, however, changes with (B,H,Ω) through T_vw in Equation (21). Section 5.3 states only that the models are driven by spatially and temporally uncorrelated inputs; it does not specify how B and H were selected from Z, nor why that selection is privileged. Section 4.4 explicitly says that it should be independently considered whether the constructed colored-in-time forcing models preserve important aspects of the original linearized dynamics, but the paper does not supply that analysis. Please specify the factorization and Ω used in Figure 11, justify the choice (e.g., via the minimal-realization construction in Ref. (62)), and either add a sensitivity analysis over feasible realizations or clearly label the spectral comparison as an illustrative completion rather than a validated prediction.
  2. [§5.1, Fig. 8] The independent validation of two-point statistics rests on a single scalar metric, the relative Frobenius norm ||V - V_DNS||_F / ||V_DNS||_F ≈ 0.6, evaluated at one wavenumber pair and γ = 300. Because the one-point correlations are enforced by equality constraints, the off-diagonal two-point recovery is the main predictive evidence, and a 60% recovery is modest. The text should report per-component or banded errors, provide a baseline comparison (e.g., white-in-time forcing or the eddy-viscosity-enhanced linearized model), and indicate the sensitivity of the metric to γ and k, or temper the claim of high-quality recovery in Section 5.1.
minor comments (6)
  1. [§5.3] The definition Π_v(k,ω) = trace(T_vw(k,ω) T_vw*(k,ω)) implicitly assumes Ω = I; if the white-noise covariance is not normalized, the output PSD should include Ω. Please state the assumed Ω explicitly.
  2. [§5.2 and §5.3] The rank-6 statement is made for γ = 10^4 in Figure 9(a), while the spectral comparison in Figure 11(a) uses γ = 300; the text should explain the relationship between these two regularization levels and the resulting rank and spectral properties.
  3. [§5.1, Fig. 7] The exact reproduction of the one-point correlation profiles is a consistency check, since those entries are imposed as equality constraints in CC-1; the text should state explicitly that the predictive content lies in the two-point and spatio-temporal statistics.
  4. [Eq. (14)] The formula R_vv(k,τ) = C(k) X(k) e^{A*(k) τ} C*(k) is valid for τ ≥ 0; for negative τ the appropriate adjoint expression should be stated.
  5. [Future Issues 1] There is a typo: 'well-possed' should be 'well-posed'.
  6. [Title page] The DOI placeholder 'https://doi.org/10.1146/((please add article doi))' must be completed before publication.

Circularity Check

1 steps flagged · score 3.0 of 10

One-point covariance match is enforced by CC-1 constraints, but two-point and spectral comparisons are independent; the spectral model is underdetermined rather than circular.

  1. self definitional [Section 5.1, Eq. (17c) and Figure 7]
    "As demonstrated in (56), optimization problem CC-1 is feasible at all wavenumbers k. Thus, regardless of the value of the regularization parameter γ, all available one-point correlations of turbulent channel flow can be reproduced by a stochastically-forced linearized model. Figure 7 displays perfect matching of all one-point velocity correlations that result from integration over wall-parallel wavenumbers."

    The available one-point correlations are exactly the entries imposed on CC-1 through the equality constraint (CXC*)ij = Vij for (i,j) in I, Eq. (17c). Any feasible solution of CC-1 satisfies those equalities by construction, so the perfect match in Figure 7 is a consistency check with the optimization's own data constraints rather than an independent prediction. The paper is transparent in calling this 'reproduced' rather than predicted, and the off-diagonal two-point correlations and the temporal spectrum are not part of the constraint set, which limits the circularity to this one demonstration.

full rationale

The main circular step is localized: CC-1 is defined with hard equality constraints on the available one-point output-covariance entries, so Figure 7's exact agreement is enforced, not predicted. The paper's more substantive claims have independent content. The off-diagonal two-point covariance entries in Figure 8 are not used as data in CC-1, and the reported ~60% Frobenius-norm recovery is therefore a genuine predictive test. The spatio-temporal PSD in Section 5.3 is also not constrained by CC-1, so comparing it with DNS is an independent check; however, Section 4.2.1 leaves the factorization Z = BH* + HB* and the white-noise covariance Ω unspecified, so the PSD is not uniquely determined by the fitted covariance and depends on additional modeling choices. The paper itself flags this limitation in Section 4.4: 'it should be independently considered whether the so-constructed colored-in-time forcing models preserve important aspects of the original linearized NS dynamics.' This is underdetermination and a validation gap, not circularity. Self-citations to (56) and (62) support feasibility and filter construction, but the necessary rank condition and Lyapunov identities are stated in the paper itself, so the citations are not the sole load-bearing evidence. Overall, the central completion claims are not forced by the input data by construction, and the circularity is confined to the one-point reproduction demonstration.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The framework relies on the linearized NS operator being stable, on stationarity and ergodicity, on the hypothesis that nonlinear terms act as additive stationary forcing, and on hard equality constraints from DNS data. The free parameters are the regularization weight γ, the noise covariance Ω, and the non-unique factorization of Z. No new physical entities are introduced; colored-in-time forcing is a standard modeling abstraction.

free parameters (3)
  • Regularization parameter γ in CC-1 = γ=300 for two-point recovery; γ=10^4 for stochastic simulations
    γ trades maximum entropy against the nuclear norm of Z. The paper chooses different values for different results and states that recovery quality depends on γ (Sections 5.1 and 5.2), but no data-driven selection rule or cross-validation is provided.
  • White-noise covariance Ω of the filter input w = Spatially and temporally uncorrelated (identity-like) in Section 5.3
    The filter realization in Equation 18 requires specifying Ω, which directly shapes Sdd and therefore the predicted spatio-temporal spectrum. The paper assumes uncorrelated white noise for all models but does not infer Ω from data or test alternatives.
  • Factorization Z = B H* + H B* = A particular B and H chosen per wavenumber, not unique
    Section 4.2.1 constructs the filter from 'matrices B(k) and H(k) that correspond to the factorization'. Different factorizations leave the fitted covariance X unchanged but alter K in Equation 19 and the temporal spectrum, so the spectral prediction depends on this unstated degree of freedom.
assumptions (5)
  • domain assumption The linearized NS equations around the turbulent mean velocity profile are stable, so all eigenvalues of A lie in the left-half plane.
    Invoked in Section 3 and Remark 1 to justify steady-state covariances and Lyapunov equations; cited to refs 57 and 58.
  • domain assumption Turbulent flow is statistically stationary and ergodic, so time averages equal ensemble expectations.
    Used throughout Sections 2 and 3 to define mean quantities, covariances, and spectra.
  • domain assumption The effect of the nonlinear terms in the Navier-Stokes equations can be represented by additive stationary stochastic forcing on the linearized dynamics.
    This is the core modeling hypothesis introduced in Sections 1, 3, and 4. It is not derived from first principles and is the foundation of the entire framework.
  • domain assumption The DNS-derived one-point correlations used in CC-1 are exact, and the equality constraints are hard constraints without a measurement noise model.
    Section 5.1 treats the available correlations as exact problem data, so any error in the DNS statistics directly propagates into the inferred covariance and forcing model.
  • standard math Standard results in linear systems theory and convex analysis: controllability, Lyapunov equations, and convexity of semidefinite programs.
    The admissibility conditions in the sidebar and the formulation of CC-1 and CC-2 rely on these classical results, cited to refs 59, 60, and 64.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Stochastic dynamical modeling of turbulent flows." pith.science (2026). https://pith.science/paper/3IV3DTWO

@misc{pith2026190809487,
  author       = {Pith},
  title        = {Pith review of: Stochastic dynamical modeling of turbulent flows},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/3IV3DTWO}},
  note         = {Machine review of arXiv:1908.09487}
}
read the original abstract

Advanced measurement techniques and high performance computing have made large data sets available for a wide range of turbulent flows that arise in engineering applications. Drawing on this abundance of data, dynamical models can be constructed to reproduce structural and statistical features of turbulent flows, opening the way to the design of effective model-based flow control strategies. This review describes a framework for completing second-order statistics of turbulent flows by models that are based on the Navier-Stokes equations linearized around the turbulent mean velocity. Systems theory and convex optimization are combined to address the inherent uncertainty in the dynamics and the statistics of the flow by seeking a suitable parsimonious correction to the prior linearized model. Specifically, dynamical couplings between states of the linearized model dictate structural constraints on the statistics of flow fluctuations. Thence, colored-in-time stochastic forcing that drives the linearized model is sought to account for and reconcile dynamics with available data (i.e., partially known second order statistics). The number of dynamical degrees of freedom that are directly affected by stochastic excitation is minimized as a measure of model parsimony. The spectral content of the resulting colored-in-time stochastic contribution can alternatively be seen to arise from a low-rank structural perturbation of the linearized dynamical generator, pointing to suitable dynamical corrections that may account for the absence of the nonlinear interactions in the linearized model.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

121 extracted references · 80 canonical work pages

  1. [1]

    Joslin RD. 1998. Aircraft laminar flow control. Ann. Rev. Fluid Mech. 30:1–29

    1998

  2. [2]

    Gad-el Hak M. 2000. Flow control: passive, active, and reactive flow management. New York: Cambridge University Press

    2000

  3. [3]

    Choi H, Moin P. 2012. Grid-point requirements for large eddy simulation: Chapman’s estimates revisited. Phys. Fluids 24:011702

    2012

  4. [4]

    Slotnick J, Khodadoust A, Alonso J, Darmofal D, Gropp W, et al. 2014. CFD vision 2030 study: A path to revolutionary computational aerosciences. Tech. Rep. CR-2014-218178, NASA

    2014

  5. [5]

    Sagaut P. 2006. Large eddy simulation for incompressible flows: an introduction. Springer Science & Business Media

    2006

  6. [6]

    Wilcox DC. 1998. Turbulence modeling for CFD, vol. 2. DCW industries La Canada, CA

    1998

  7. [7]

    Durbin PA, Reif BAP. 2011. Statistical theory and modeling for turbulent flows. Wiley

    2011

  8. [8]

    Kim J, Bewley TR. 2007. A linear systems approach to flow control. Annu. Rev. Fluid Mech. 39:383–417

    2007

Show all 121 references
  1. [9]

    Robinson SK. 1991. Coherent motions in the turbulent boundary layer. Annu. Rev. Fluid Mech. 23:601–639

    1991

  2. [10]

    Adrian RJ. 2007. Hairpin vortex organization in wall turbulence. Phys. Fluids 19:041301

    2007

  3. [11]

    Smits AJ, McKeon BJ, Marusic I. 2011. High-Reynolds number wall turbulence. Ann. Rev. Fluid Mech. 43:353–375

    2011

  4. [12]

    Jim´ enez J. 2018. Coherent structures in wall-bounded turbulence. J. Fluid Mech. 842

    2018

  5. [13]

    Rowley CW. 2005. Model reduction for fluids using balanced proper orthogonal decomposition. Intl J. Bifurcation Chaos 15:997–1013

    2005

  6. [14]

    Lumley JL. 2007. Stochastic tools in turbulence. Dover Publications www.annualreviews.org • Stochastic dynamical modeling of turbulent flows 21

    2007

  7. [15]

    Schmid PJ. 2010. Dynamic mode decomposition of numerical and experimental data. J. Fluid Mech. 656:5–28

    2010

  8. [16]

    Jovanovi´ c MR, Schmid PJ, Nichols JW. 2014. Sparsity-promoting dynamic mode decomposi- tion. Phys. Fluids 26:024103 (22 pages)

    2014

  9. [17]

    Rowley CW, Dawson ST. 2017. Model reduction for flow analysis and control. Annu. Rev. Fluid Mech. 49:387–417

    2017

  10. [18]

    Towne A, Schmidt OT, Colonius T. 2018. Spectral proper orthogonal decomposition and its relationship to dynamic mode decomposition and resolvent analysis. J. Fluid Mech. 847:821– 867

    2018

  11. [19]

    Noack BR, Morzy´ nski M, Tadmor G. 2011. Reduced-order modelling for flow control, vol. 528 of CISM Courses and Lectures . Springer

    2011

  12. [20]

    Tadmor G, Noack BR. 2011. Bernoulli, Bode, and Budgie [Ask the Experts]. IEEE Contr. Syst. Mag. 31:18–23

    2011

  13. [21]

    Trefethen LN, Trefethen AE, Reddy SC, Driscoll TA. 1993. Hydrodynamic stability without eigenvalues. Science 261:578–584

    1993

  14. [22]

    Schmid PJ. 2007. Nonmodal stability theory. Annu. Rev. Fluid Mech. 39:129–162

    2007

  15. [23]

    Gustavsson LH. 1991. Energy growth of three-dimensional disturbances in plane Poiseuille flow. J. Fluid Mech. 98:149

    1991

  16. [24]

    Butler KM, Farrell BF. 1992. Three-dimensional optimal perturbations in viscous shear flow. Phys. Fluids A 4:1637

    1992

  17. [25]

    Reddy SC, Henningson DS. 1993. Energy growth in viscous channel flows. J. Fluid Mech. 252:209–238

    1993

  18. [26]

    Henningson DS, Reddy SC. 1994. On the role of linear mechanisms in transition to turbulence. Phys. Fluids 6:1396–1398

    1994

  19. [27]

    Schmid PJ, Henningson DS. 1994. Optimal energy density growth in Hagen-Poiseuille flow. J. Fluid Mech. 277:197–225

    1994

  20. [28]

    Farrell BF, Ioannou PJ. 1993. Stochastic forcing of the linearized Navier-Stokes equations. Phys. Fluids A 5:2600–2609

    1993

  21. [29]

    Bamieh B, Dahleh M. 2001. Energy amplification in channel flows with stochastic excitation. Phys. Fluids 13:3258–3269

    2001

  22. [30]

    Jovanovi´ c MR. 2004. Modeling, analysis, and control of spatially distributed systems. Ph.D. thesis, University of California, Santa Barbara

    2004

  23. [31]

    Jovanovi´ c MR, Bamieh B. 2005. Componentwise energy amplification in channel flows. J. Fluid Mech. 534:145–183

    2005

  24. [32]

    Ran W, Zare A, Hack MJP, Jovanovi´ c MR. 2019. Stochastic receptivity analysis of boundary layer flow. Phys. Rev. Fluids Doi:10.1103/PhysRevFluids.00.003900; also arXiv:1807.07759

    2019 arXiv

  25. [33]

    Butler KM, Farrell BF. 1993. Optimal perturbations and streak spacing in wall-bounded tur- bulent shear flow. Phys. Fluids A 5:774–777

    1993

  26. [34]

    Farrell BF, Ioannou PJ. 1993. Optimal excitation of three-dimensional perturbations in viscous constant shear flow. Phys. Fluids A 5:1390–1400

    1993

  27. [35]

    Farrell BF, Ioannou PJ. 1998. Perturbation structure and spectra in turbulent channel flow. Theoret. Comput. Fluid Dynamics 11:237–250

    1998

  28. [36]

    McKeon BJ, Sharma AS. 2010. A critical-layer framework for turbulent pipe flow. J. Fluid Mech. 658:336–382

    2010

  29. [37]

    Moarref R, Sharma AS, Tropp JA, McKeon BJ. 2013. Model-based scaling of the streamwise energy density in high-Reynolds-number turbulent channels. J. Fluid Mech. 734:275–316

    2013

  30. [38]

    Moarref R, Jovanovi´ c MR, Tropp JA, Sharma AS, McKeon BJ. 2014. A low-order decompo- sition of turbulent channel flow via resolvent analysis and convex optimization. Phys. Fluids 26:051701 (7 pages)

    2014

  31. [39]

    McComb WD. 1991. The physics of fluid turbulence. Oxford University Press

    1991

  32. [40]

    Kraichnan RH. 1959. The structure of isotropic turbulence at very high Reynolds numbers. J. 22 Zare, Georgiou, Jovanovi´ c Fluid Mech. 5:497–543

    1959

  33. [41]

    Kraichnan RH. 1971. An almost-Markovian Galilean-invariant turbulence model. J. Fluid Mech. 47:513–524

    1971

  34. [42]

    Orszag SA. 1970. Analytical theories of turbulence. J. Fluid Mech. 41:363–386

    1970

  35. [43]

    Monin AS, Yaglom AM. 1975. Statistical fluid mechanics, vol. 2. MIT Press

    1975

  36. [44]

    Farrell BF, Ioannou PJ. 1993. Stochastic dynamics of baroclinic waves.J. Atmos. Sci. 50:4044– 4057

    1993

  37. [45]

    Farrell BF, Ioannou PJ. 1994. A theory for the statistical equilibrium energy spectrum and heat flux produced by transient baroclinic waves. J. Atmos. Sci. 51:2685–2698

    1994

  38. [46]

    DelSole T, Farrell BF. 1995. A stochastically excited linear system as a model for quasi- geostrophic turbulence: Analytic results for one-and two-layer fluids. J. Atmos. Sci. 52:2531– 2547

    1995

  39. [47]

    Hwang Y, Cossu C. 2010. Amplification of coherent streaks in the turbulent Couette flow: an input-output analysis at low Reynolds number. J. Fluid Mech. 643:333–348

    2010

  40. [48]

    Hwang Y, Cossu C. 2010. Linear non-normal energy amplification of harmonic and stochastic forcing in the turbulent channel flow. J. Fluid Mech. 664:51–73

    2010

  41. [49]

    Jovanovi´ c MR, Bamieh B. 2001. Modelling flow statistics using the linearized Navier-Stokes equations. In Proceedings of the 40th IEEE Conference on Decision and Control

    2001

  42. [50]

    Durbin PA, Reif BAP. 2000. Theory and modeling of turbulent flows. Wiley

    2000

  43. [51]

    Pope SB. 2000. Turbulent flows. Cambridge University Press

    2000

  44. [52]

    Jones W, Launder B. 1972. The prediction of laminarization with a two-equation model of turbulence. Int. J. Heat Mass Transf. 15:301–314

    1972

  45. [53]

    Launder B, Sharma B. 1974. Application of the energy-dissipation model of turbulence to the calculation of flow near a spinning disc. Lett. Heat Mass Transf. 1:131–137

    1974

  46. [54]

    Kim J, Moin P, Moser R. 1987. Turbulence statistics in fully developed channel flow at low Reynolds number. J. Fluid Mech. 177:133–166

    1987

  47. [55]

    Weideman JAC, Reddy SC. 2000. A MATLAB differentiation matrix suite.ACM Trans. Math. Software 26:465–519

    2000

  48. [56]

    Zare A, Jovanovi´ c MR, Georgiou TT. 2017. Colour of turbulence.J. Fluid Mech. 812:636–680

    2017

  49. [57]

    Malkus WVR. 1956. Outline of a theory of turbulent shear flow. J. Fluid Mech. 1:521–539

    1956

  50. [58]

    Reynolds WC, Tiederman WG. 1967. Stability of turbulent channel flow with application to Malkus’s theory. J. Fluid Mech. 27:253–272

    1967

  51. [59]

    Georgiou TT. 2002. Spectral analysis based on the state covariance: the maximum entropy spectrum and linear fractional parametrization. IEEE Trans. Autom. Control 47:1811–1823

    2002

  52. [60]

    Georgiou TT. 2002. The structure of state covariances and its relation to the power spectrum of the input. IEEE Trans. Autom. Control 47:1056–1066

    2002

  53. [61]

    Moin P, Moser R. 1989. Characteristic-eddy decomposition of turbulence in a channel. J. Fluid Mech. 200:509

    1989

  54. [62]

    Zare A, Chen Y, Jovanovi´ c MR, Georgiou TT. 2017. Low-complexity modeling of partially available second-order statistics: theory and an efficient matrix completion algorithm. IEEE Trans. Automat. Control 62:1368–1383

    2017

  55. [63]

    Chen Y, Jovanovi´ c MR, Georgiou TT. 2013. State covariances and the matrix completion problem. In Proceedings of the 52nd IEEE Conference on Decision and Control

    2013

  56. [64]

    Boyd S, Vandenberghe L. 2004. Convex optimization. Cambridge University Press

    2004

  57. [65]

    Goodwin GC, Payne RL. 1977. Dynamic system identification: experiment design and data analysis. Academic press

    1977

  58. [66]

    Fazel M. 2002. Matrix rank minimization with applications. Ph.D. thesis, Stanford University

    2002

  59. [67]

    Recht B, Fazel M, Parrilo PA. 2010. Guaranteed minimum-rank solutions of linear matrix equations via nuclear norm minimization. SIAM Rev. 52:471–501

    2010

  60. [68]

    Hotz A, Skelton RE. 1987. Covariance control theory. Int. J. Control 46:13–32

    1987

  61. [69]

    Yasuda K, Skelton RE, Grigoriadis KM. 1993. Covariance controllers: A new parametrization www.annualreviews.org • Stochastic dynamical modeling of turbulent flows 23 of the class of all stabilizing controllers. Automatica 29:785–788

    1993

  62. [70]

    Grigoriadis KM, Skelton RE. 1994. Alternating convex projection methods for covariance control design. Int. J. Control 60:1083–1106

    1994

  63. [71]

    Chen Y, Georgiou TT, Pavon M. 2016. Optimal steering of a linear stochastic system to a final probability distribution, Part II. IEEE Trans. Automat. Control 61:1170–1180

    2016

  64. [72]

    Lin F, Jovanovi´ c MR. 2009. Least-squares approximation of structured covariances. IEEE Trans. Automat. Control 54:1643–1648

    2009

  65. [73]

    Zorzi M, Ferrante A. 2012. On the estimation of structured covariance matrices. Automatica 48:2145–2151

    2012

  66. [74]

    Zare A, Jovanovi´ c MR, Georgiou TT. 2016. Perturbation of system dynamics and the co- variance completion problem. In Proceedings of the 55th IEEE Conference on Decision and Control

    2016

  67. [75]

    Zare A, Mohammadi H, Dhingra NK, Jovanovi´ c MR, Georgiou TT. 2018. Proximal algo- rithms for large-scale statistical modeling and optimal sensor/actuator selection. IEEE Trans. Automat. Control Submitted; also arXiv:1807.01739

    2018 arXiv

  68. [76]

    Zare A, Jovanovi´ c MR, Georgiou TT. 2015. Alternating direction optimization algorithms for covariance completion problems. In Proceedings of the 2015 American Control Conference

    2015

  69. [77]

    Toh KC, Todd MJ, T¨ ut¨ unc¨ u RH. 1999. SDPT3-a MATLAB software package for semidefinite programming, version 1.3. Optim. Methods Softw. 11:545–581

    1999

  70. [78]

    Grant M, Boyd S. 2014. CVX: Matlab software for disciplined convex programming, version 2.1. http://cvxr.com/cvx

    2014

  71. [79]

    Sasaki K, Piantanida S, Cavalieri AVG, Jordan P. 2017. Real-time modelling of wavepackets in turbulent jets. J. Fluid Mech. 821:458–481

    2017

  72. [80]

    Beneddine S, Sipp D, Arnault A, Dandois J, Lesshafft L. 2016. Conditions for validity of mean flow stability analysis. J. Fluid Mech. 798:485–504

    2016

  73. [81]

    Beneddine S, Yegavian R, Sipp D, Leclaire B. 2017. Unsteady flow dynamics reconstruction from mean flow and point sensors: an experimental study. J. Fluid Mech. 824:174–201

    2017

  74. [82]

    Towne A, Lozano-Dur´ an A, Yang X. 2019. Resolvent-based estimation of space-time flow statistics. arXiv:1901.07478

    2019 arXiv

  75. [83]

    Morra P, Semeraro O, Henningson DS, Cossu C. 2019. On the relevance of Reynolds stresses in resolvent analyses of turbulent wall-bounded flows. J. Fluid Mech. 867:969–984

    2019

  76. [84]

    Moser RD, Kim J, Mansour NN. 1999. DNS of turbulent channel flow up to Reτ = 590. Phys. Fluids 11:943–945

    1999

  77. [85]

    Del ´Alamo JC, Jim´ enez J. 2003. Spectra of the very large anisotropic scales in turbulent channels. Phys. Fluids 15:41–44

    2003

  78. [86]

    Del ´Alamo JC, Jim´ enez J, Zandonade P, Moser RD. 2004. Scaling of the energy spectra of turbulent channels. J. Fluid Mech. 500:135–144

    2004

  79. [87]

    Moarref R, Jovanovi´ c MR. 2012. Model-based design of transverse wall oscillations for turbu- lent drag reduction. J. Fluid Mech. 707:205–240

    2012

  80. [88]

    Monty JP, Stewart JA, Williams RC, Chong MS. 2007. Large-scale features in turbulent pipe and channel flows. J. Fluid Mech. 589:147–156

    2007

  81. [89]

    Reynolds WC, Hussain AKMF. 1972. The mechanics of an organized wave in turbulent shear flow. Part 3. Theoretical models and comparisons with experiments.J. Fluid Mech. 54:263–288

    1972

  82. [90]

    Del ´Alamo JC, Jim´ enez J. 2006. Linear energy amplification in turbulent channels. J. Fluid Mech. 559:205–213

    2006

  83. [91]

    Cossu C, Pujals G, Depardon S. 2009. Optimal transient growth and very large-scale structures in turbulent boundary layers. J. Fluid Mech. 619:79–94

    2009

  84. [92]

    Pujals G, Garc´ ıa-Villalba M, Cossu C, Depardon S. 2009. A note on optimal transient growth in turbulent channel flows. Phys. Fluids 21:015109

    2009

  85. [93]

    Hœpffner J, Chevalier M, Bewley TR, Henningson DS. 2005. State estimation in wall-bounded flow systems. Part 1. Perturbed laminar flows. J. Fluid Mech. 534:263–294 24 Zare, Georgiou, Jovanovi´ c

    2005

  86. [94]

    Chevalier M, Hœpffner J, Bewley TR, Henningson DS. 2006. State estimation in wall-bounded flow systems. Part 2. Turbulent flows. J. Fluid Mech. 552:167–187

    2006

  87. [95]

    Bewley TR, Liu S. 1998. Optimal and robust control and estimation of linear paths to transi- tion. J. Fluid Mech. 365:305–349

    1998

  88. [96]

    H¨ ogberg M, Bewley TR, Henningson DS. 2003. Linear feedback control and estimation of transition in plane channel flow. J. Fluid Mech. 481:149–175

    2003

  89. [97]

    Fransson JHM, Talamelli A, Brandt L, Cossu C. 2006. Delaying transition to turbulence by a passive mechanism. Phys. Rev. Lett. 96:064501

    2006

  90. [98]

    Jovanovi´ c MR. 2008. Turbulence suppression in channel flows by small amplitude transverse wall oscillations. Phys. Fluids 20:014101 (11 pages)

    2008

  91. [99]

    Moarref R, Jovanovi´ c MR. 2010. Controlling the onset of turbulence by streamwise traveling waves. Part 1: Receptivity analysis. J. Fluid Mech. 663:70–99

    2010

  92. [100]

    Lieu BK, Moarref R, Jovanovi´ c MR. 2010. Controlling the onset of turbulence by streamwise traveling waves. Part 2: Direct numerical simulations. J. Fluid Mech. 663:100–119

    2010

  93. [101]

    Hoda N, Jovanovi´ c MR, Kumar S. 2008. Energy amplification in channel flows of viscoelastic fluids. J. Fluid Mech. 601:407–424

    2008

  94. [102]

    Hoda N, Jovanovi´ c MR, Kumar S. 2009. Frequency responses of streamwise-constant pertur- bations in channel flows of Oldroyd-B fluids. J. Fluid Mech. 625:411–434

    2009

  95. [103]

    Jovanovi´ c MR, Kumar S. 2010. Transient growth without inertia.Phys. Fluids 22:023101 (19 pages)

    2010

  96. [104]

    Jovanovi´ c MR, Kumar S. 2011. Nonmodal amplification of stochastic disturbances in strongly elastic channel flows. J. Non-Newtonian Fluid Mech. 166:755–778

    2011

  97. [105]

    Lieu BK, Jovanovi´ c MR, Kumar S. 2013. Worst-case amplification of disturbances in inertialess Couette flow of viscoelastic fluids. J. Fluid Mech. 723:232–263

    2013

  98. [106]

    Jeun J, Nichols JW, Jovanovi´ c MR. 2016. Input-output analysis of high-speed axisymmetric isothermal jet noise. Phys. Fluids 28:047101 (20 pages)

    2016

  99. [107]

    Hildebrand N, Dwivedi A, Nichols JW, Jovanovi´ c MR, Candler GV. 2018. Simulation and stability analysis of oblique shock wave/boundary layer interactions at Mach 5.92. Phys. Rev. Fluids 3:013906 (23 pages)

    2018

  100. [108]

    Dwivedi A, Sidharth GS, Nichols JW, Candler GV, Jovanovi´ c MR. 2019. Reattachment vor- tices in hypersonic compression ramp flow: an input-output analysis.J. Fluid Mech. To appear; also arXiv:1811.09046

    2019 arXiv

  101. [109]

    Reed HL, Saric WS, Arnal D. 1996. Linear stability theory applied to boundary layers. Annu. Rev. Fluid Mech. 28:389–428

    1996

  102. [110]

    Herbert T. 1997. Parabolized stability equations. Annu. Rev. Fluid Mech. 29:245–283

    1997

  103. [111]

    H¨ ogberg M, Henningson DS. 2002. Linear optimal control applied to instabilities in spatially developing boundary layers. J. Fluid Mech. 470:151–179

    2002

  104. [112]

    Ran W, Zare A, Hack MJP, Jovanovi´ c MR. 2019. Modeling mode interactions in boundary layer flows via Parabolized Floquet Equations. Phys. Rev. Fluids 4:023901 (22 pages)

    2019

  105. [113]

    Cand` es EJ, Recht B. 2009. Exact matrix completion via convex optimization.Found. Comput. Math. 9:717–772

    2009

  106. [114]

    Absil PA, Mahony R, Sepulchre R. 2008. Optimization algorithms on matrix manifolds. Prince- ton, NJ: Princeton University Press

    2008

  107. [115]

    Grussler C, Zare A, Jovanovi´ c MR, Rantzer A. 2016. The use of ther∗ heuristic in covariance completion problems. In Proceedings of the 55th IEEE Conference on Decision and Control

    2016

  108. [116]

    Grussler C, Rantzer A, Giselsson P. 2018. Low-rank optimization with convex constraints. IEEE Trans. Automat. Control

    2018

  109. [117]

    Candes EJ, Li X, Soltanolkotabi M. 2015. Phase retrieval via Wirtinger flow: Theory and algorithms. IEEE Trans. Inf. Theory 61:1985–2007

    2015

  110. [118]

    Sun R, Luo ZQ. 2016. Guaranteed matrix completion via non-convex factorization. IEEE Trans. Inf. Theory 62:6535–6579 www.annualreviews.org • Stochastic dynamical modeling of turbulent flows 25

    2016

  111. [119]

    Ge R, Lee JD, Ma T. 2016. Matrix completion has no spurious local minimum. In Adv. Neural Inf. Process Syst

    2016

  112. [120]

    Karabasov SA, Afsar MZ, Hynes TP, Dowling AP, McMullan WA, et al. 2010. Jet noise: acoustic analogy informed by large eddy simulation. AIAA Journal 48:1312–1325

    2010

  113. [121]

    Leib SJ, Goldstein ME. 2011. Hybrid source model for predicting high-speed jet noise. AIAA Journal 49:1324–1335 26 Zare, Georgiou, Jovanovi´ c

    2011

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.