Abstract
The manuscript deals with continuous adjoint companions of prominent explicit Large Eddy Simulation (LES) methods grounding on the eddy viscosity assumption for incompressible fluids. The subgrid-scale approximations considered herein address the classic Smagorinsky-Lilly, the Wall-Adapting Local Eddy-Viscosity (WALE), and the Kinetic Energy Subgrid-Scale (KESS) model, whereby only static implementations, i.e., those without dynamically adjusted model parameters, are considered. The associated continuous adjoint systems and resulting shape sensitivity expressions are derived. Information on the consistent discrete implementation is provided that benefits from the self-adjoint primal discretization of convective and diffusive fluxes via unbiased, symmetric approximations, frequently performed in explicit LES studies to minimize numerical diffusion. Algebraic primal subgrid-scale models yield algebraic adjoint LES relationships that resemble additional adjoint momentum sources. The KESS one equation model introduces an additional adjoint equation, which enlarges the resulting continuous adjoint KESS system with potentially increased numerical stiffness. The different adjoint LES methods are tested and compared against each other on a flow around a circular cylinder at \(\text{Re}_\text{D} = {140000\,}\) for a boundary (drag) and volume (deviation from target velocity distribution) based cost functional. Since all primal implementations predict similar flow fields, it is possible to swap the associated adjoint systems –i.e., applying an adjoint WALE method to a primal KESS result– and still obtain plausible adjoint results. Due to the LES’s inherent unsteady character, the adjoint solver requires the entire primal flow field over the cost-functional relevant time horizon. Even for the academic cases studied herein, the storage capacities are in the order of terabytes and refer to a practical bottleneck. However, in the case of suitable, time-averaged cost functional, the time-averaged primal flow field can be used directly in a steady adjoint solver, which results in a drastic effort reduction.
Similar content being viewed by others
Data Availability
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
Achenbach, E.: Distribution of local pressure and skin friction around a circular cylinder in cross-flow up to Re= \(5\times 10^6\). Journal of Fluid Mechanics 34(4), 625–639 (1968). https://doi.org/10.1017/S0022112068002120
Angerbauer, R., Rung, T.: Hybrid RANS/LES Simulations of Aerodynamic Flows Around Superstructures of Ships. In: 17th Symposium on Hybrid RANS-LES Methods, 17–19 September, 2018, Berlin, Germany, pages 367–377. Springer (2020). https://doi.org/10.1007/978-3-030-27607-2_30
ANSYS Inc. Ansys Fluent Theory Guide, Release 2021 R2, Cha. 4.16, Large Eddy Simulation (LES) Model (2021)
Asmuth, H., Korb, H., and Ivanell, S.: How Fast is Fast Enough? Industry Perspectives on the Use of Large-eddy Simulation in Wind Energy. In: Journal of Physics: Conference Series, volume 2505, page 012001. IOP Publishing (2023).https://doi.org/10.1088/1742-6596/2505/1/012001
Belyaev, A., Fayolle, P.-A.: An ADMM-based scheme for distance function approximation. Numerical Algorithms 84, 983–996 (2020). https://doi.org/10.1007/s11075-019-00789-5
Belyaev, A.G., Fayolle, P.A.: On variational and PDE-based distance function approximations. Computer Graphics Forum 34(8), 104–118 (2015). https://doi.org/10.1111/cgf.12611
Bewley, T., Choi, H., Temam, R., and Moin, P.: Optimal Feedback Control of Turbulent Channel Flow. Annual Research Briefs (1993)
Bletzinger, K.-U.: A consistent frame for sensitivity filtering and the vertex assigned morphing of optimal shape. Structural and Multidisciplinary Optimization 49(6), 873–895 (2014). https://doi.org/10.1007/s00158-013-1031-5
Boris, J.P., Grinstein, F.F., Oran, E.S., Kolbe, R.L.: New insights into large Eddy simulation. Fluid Dynamics Research 10(4–6), 199 (1992). https://doi.org/10.1016/0169-5983(92)90023-P
Bouffanais, R.: Advances and challenges of applied large-Eddy simulation. Computers & Fluids 39(5), 735–738 (2010). https://doi.org/10.1016/j.compfluid.2009.12.003
Boyd, S.P., Vandenberghe, L.: Convex optimization. Cambridge University Press (2004)
Breuer, M., Simulation, Direkte Numerische, Simulation, Large-Eddy., turbulenter Strömungen auf Hochleistungsrechnern. Shaker Verlag, Aachen,: Berichte aus der Strömungstechnik. Universität Erlangen-Nürnberg, Habil.-Schr (2002)
Cantwell, B., Coles, D.: An experimental study of entrainment and transport in the turbulent near wake of a circular cylinder. Journal of Fluid Mechanics 136, 321–374 (1983). https://doi.org/10.1017/S0022112083002189
Choi, H., Temam, R., Moin, P., Kim, J.: Feedback control for unsteady flow and its application to the stochastic burgers equation. Journal of Fluid Mechanics 253, 509–543 (1993). https://doi.org/10.1017/S0022112093001880
Davidson, L.: How to Estimate the Resolution of an LES of Recirculating Flow. In Quality and Reliability of Large-Eddy Simulations II, pages 269–286. Springer (2011). https://doi.org/10.1007/978-94-007-0231-8_25
Ducros, F., Nicoud, F., Poinsot, T.: Wall-adapting local Eddy-viscosity models for simulations in complex geometries. Numerical Methods for Fluid Dynamics VI 6, 293–299 (1998)
Fares, E., Schröder, W.: A differential equation for approximate wall distance. International journal for numerical methods in fluids 39(8), 743–762 (2002). https://doi.org/10.1002/fld.348
Ferziger, J.H., Peric, M.: Computational methods for fluid dynamics. Springer (2012)
Fröhlich, J., Rodi, W.: Introduction to large Eddy simulation of turbulent flows. University of Karlsruhe (2002)
Fureby, C., Grinstein, F.F.: Monotonically integrated large eddy simulation of free shear flows. AIAA Journal 37(5), 544–556 (1999). https://doi.org/10.2514/2.772
Gant, S.E.: Reliability issues of LES-related approaches in an industrial context. Flow, Turbulence and Combustion 84, 325–335 (2010). https://doi.org/10.1007/s10494-009-9237-8
Georgiadis, N.J., Rizzetta, D.P., Fureby, C.: Large-Eddy simulation: current capabilities, recommended practices, and future research. AIAA journal 48(8), 1772–1784 (2010). https://doi.org/10.2514/1.J050232
Gerdes, A.: Fluid Dynamic Optimization of HVAC-Components with Adjoint Methods. PhD thesis, Hamburg University of Technology (2018)
Gerdes, A., Hinze, M., Rung, T.: An efficient line search technique and its application to adjoint topology optimisation. PAMM 14(1), 719–720 (2014). https://doi.org/10.1002/pamm.201410342
Germano, M., Piomelli, U., Moin, P., Cabot, W.H.: A dynamic subgrid-scale Eddy viscosity model. Physics of Fluids A: Fluid Dynamics 3(7), 1760–1765 (1991). https://doi.org/10.1063/1.857955
Giles, M.B., Pierce, N.A.: Adjoint equations in CFD: duality, boundary conditions and solution behaviour. AIAA Paper (1997). https://doi.org/10.2514/6.1997-1850.AIAA-97-1850
Griewank, A., Walther, A.: Revolve - reverse or adjoint mode of computational differentiation. Transaction on Mathematical Software 26(1), 19–45 (2000)
Haubner, J., Siebenborn, M., Ulbrich, M.: A continuous perspective on modeling of shape optimal design problems. SIAM Journal on Scientific Computing 43(3), A1997–A2018 (2021). https://doi.org/10.1137/20M1332050
Hill, D.C.: Adjoint systems and their role in the receptivity problem for boundary layers. Journal of Fluid Mechanics 292, 183–204 (1995). https://doi.org/10.1017/S0022112095001480
Huber, P.J.: Robust Estimation of a Location Parameter. In Breakthroughs in Statistics: Methodology and Distribution, pages 492–518. Springer (1992). https://doi.org/10.1007/978-1-4612-4380-9_35
Kim, S.-E.: Large Eddy Simulation Using an Unstructured Mesh Based Finite-Volume Solver. In: 34th AIAA Fluid Dynamics Conference and Exhibit, Portland, Oregon, AIAA–2004–2548 (2004).https://doi.org/10.2514/6.2004-2548
Kim, W.-W., and Menon, S.: Application of the Localized Dynamic Subgrid-Scale Model to Turbulent Wall-Bounded Flows. In: 35th Aerospace Sciences Meeting and Exhibit, Reno, NV, AIAA PAPER 97-0210 (1997). https://doi.org/10.2514/6.1997-210
Kröger, J., Rung, T.: CAD-free hydrodynamic optimisation using consistent kernel-based sensitivity filtering. Ship Technology Research 62(3), 111–130 (2016). https://doi.org/10.1080/09377255.2015.1109872
Kröger, J., Kühl, N., Rung, T.: Adjoint volume-of-fluid approaches for the hydrodynamic optimisation of ships. Ship Technology Research 65(1), 47–68 (2018). https://doi.org/10.1080/09377255.2017.1411001
Kühl, N.: Adjoint-Based Shape Optimization Constraint by Turbulent Two-Phase Navier-Stokes Systems. PhD thesis, Hamburg University of Technology (2021)
Kühl, N., Rung, T.: Discrete adjoint momentum-weighted interpolation strategies. Journal of Computational Physics 467, 111474 (2022). https://doi.org/10.1016/j.jcp.2022.111474
Kühl, N., Kröger, J., Siebenborn, M., Hinze, M., Rung, T.: Adjoint complement to the volume-of-fluid method for immiscible flows. Journal of Computational Physics 440, 110411 (2021). https://doi.org/10.1016/j.jcp.2021.110411
Kühl, N., Müller, P.M., Rung, T.: Adjoint complement to the universal momentum law of the wall. Flow Turbulence and Combustion (2021). https://doi.org/10.1007/s10494-021-00286-7
Kühl, N., Nguyen, T.T., Palm, M., Jürgens, D., Rung, T.: Adjoint node-based shape optimization of free floating vessels. Structural and Multidisciplinary Optimization 65, 247 (2022). https://doi.org/10.1007/s00158-022-03338-2
Kühl, N., Fischer, H., Hinze, M., Rung, T.: An incremental singular value decomposition approach for large-scale spatially parallel & distributed but temporally serial data - applied to technical flows. Computer Physics Communications 296, 109022 (2024). https://doi.org/10.1016/j.cpc.2023.109022
Larsson, J., Kawai, S., Bodart, J., Bermejo-Moreno, I.: Large Eddy simulation with modeled wall-stress: recent progress and future directions. Mechanical Engineering Reviews 3(1), 15–00418 (2016). https://doi.org/10.1299/mer.15-00418
Li, X., Hulshoff, S., Hickel, S.: Towards adjoint-based mesh refinement for large Eddy simulation using reduced-order primal solutions: preliminary 1D Burgers study. Computer Methods in Applied Mechanics and Engineering 379, 113733 (2021). https://doi.org/10.1016/j.cma.2021.113733
Lilly, D.K.: The Representation of Small-Scale Turbulence in Numerical Simulation Experiments. In: Proc. IBM Sci. Comput. Symp. on Environmental Science, pages 195–210. https://doi.org/10.5065/D62R3PMM (1967)
Löhner, R., Corrigan, A.T., Wichmann, K.R., and Wall, W.: On the Achievable Speeds of Finite Difference Solvers on CPUs and GPUs. In: 21st AIAA Computational Fluid Dynamics Conference, San Diego, AIAA 2013-2852. https://doi.org/10.2514/6.2013-2852 (2013)
Löhner, R., Corrigan, A.T., Wichmann, K.-R., and Wall, W.: Comparison of Lattice-Boltzmann and Finite Difference Solvers. In: 52nd Aerospace Sciences Meeting, National Harbor, Maryland, AIAA 2014-1439. https://doi.org/10.2514/6.2014-1439 (2014)
Löhner, R., Figueroa, A., and Degro, A.: Recent Advances in a Cartesian Solver for Industrial LES. In: AIAA Science and Technology Forum and Exposition, San Diego, CA, pages AIAA 2019–2328. https://doi.org/10.2514/6.2019-2328 (2019)
Löhner, R., Othmer, C., Mrosek, M., Figueroa, A., Degro, A.: Overnight industrial LES for external aerodynamics. Computers & Fluids 214, 104771 (2021). https://doi.org/10.1016/j.compfluid.2020.104771
Manzke, M.: Development of a Scalable Method for the Efficient Simulation of Flows using Dynamic Goal-Orientated Local Grid-Adaption. PhD thesis, Hamburg University of Technology (2018)
Mehta, D., Van Zuijlen, A.H., Koren, B., Holierhoek, J.G., Bijl, H.: Large Eddy simulation of wind farm aerodynamics: a review. Journal of Wind Engineering and Industrial Aerodynamics 133, 1–17 (2014). https://doi.org/10.1016/j.jweia.2014.07.002
Menon, S., Yeung, P.-K., Kim, W.-W.: Effect of subgrid models on the computed interscale energy transfer in isotropic turbulence. Computers & Fluids 25(2), 165–180 (1996). https://doi.org/10.1016/0045-7930(95)00036-4
Menter, F.R.: Two-Equation Eddy-viscosity turbulence models for engineering applications. AIAA Journal 32(8), 1598–1605 (1994). https://doi.org/10.2514/3.12149
Moin, P., and Bewley, T.: Feedback Control of Turbulence. Technical report, Mechanical Engineering Department, Stanford University (1994)
Nadarajah, S. K.: The Discrete Adjoint Approach to Aerodynamic Shape Optimization. PhD thesis, Stanford University (2003)
Nicoud, F., Ducros, F.: Subgrid-scale stress modelling based on the square of the velocity gradient tensor. Flow, Turbulence and Combustion 62(3), 183–200 (1999). https://doi.org/10.1023/A:1009995426001
Othmer, C.: A continuous adjoint formulation for the computation of topological and surface sensitivities of ducted flows. International Journal for Numerical Methods in Fluids 58(8), 861–877 (2008). https://doi.org/10.1002/fld.1770
Pache, R.: Datenbasierte Modellierung der Aerodynamischen Lasten auf Überwasserschiffe. PhD thesis, Hamburg University of Technology (2023)
Piomelli, U.: Large Eddy Simulations in 2030 and Beyond. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences 372(2022), 20130320 (2014). https://doi.org/10.1098/rsta.2013.0320
Piomelli, U., Zang, T.A., Speziale, C.G., Hussaini, M.Y.: On the Large-Eddy simulation of transitional wall-bounded flows. Physics of Fluids A: Fluid Dynamics 2(2), 257–265 (1990). https://doi.org/10.1063/1.857774
Pope, S.B.: Turbulent Flows. Cambridge University Press (2001)
Rhie, C.M., Chow, W.L.: Numerical Study of the turbulent flow past an airfoil with trailing edge separation. AIAA Journal 21(11), 1525–1532 (1983). https://doi.org/10.2514/3.8284
Roth, R., Ulbrich, S.: A discrete adjoint approach for the optimization of unsteady turbulent flows. Flow, Turbulence and Combustion 90, 763–783 (2013). https://doi.org/10.1007/s10494-012-9439-3
Rung, T., Wöckner, K., Manzke, M., Brunswig, J., Ulrich, C., Stück, A.: Challenges and perspectives for maritime CFD applications. Jahrbuch der Schiffbautechnischen Gesellschaft 103, 127–39 (2009)
Sagaut, P.: Large Eddy simulation for incompressible flows: an introduction. Springer (2005)
Schubert, S.: Analysis of Coupling Techniques for Overset-Grid Finite-Volume Methods. PhD thesis, Hamburg University of Technology (2019)
Siemens Digital Industries Software. Simcenter STAR-CCM+ User Guide v. 2023.6, Cha. Scale-Resolving Simulations (2023)
Smagorinsky, J.: General circulation experiments with the primitive equations: I. The Basic Experiment. Monthly Weather Review 91(3), 99–164 (1963)
Son, J.S., Hanratty, T.J.: Velocity gradients at the wall for flow around a cylinder at Reynolds numbers from \(5\times 10^3 to 10^5\). Journal of Fluid Mechanics 35(2), 353–368 (1969). https://doi.org/10.1017/S0022112069001157
Stück, A.: Adjoint Navier-Stokes Methods for Hydrodynamic Shape Optimisation. PhD thesis, Hamburg University of Technology (2012)
Stück, A., Rung, T.: Adjoint complement to viscous finite-volume pressure-correction methods. Journal of Computational Physics 248, 402–419 (2013). https://doi.org/10.1016/j.jcp.2013.01.002
Stück, A., Camelli, F.F., Löhner, R.: Adjoint-based design of shock mitigation devices. International Journal for Numerical Methods in Fluids 64(4), 443–472 (2010). https://doi.org/10.1002/fld.2164
The HDF Group: Hierarchical Data Format, Version 5. -NNNN. URL https://www.hdfgroup.org/HDF5/ (1997)
Tucker, P.G.: Differential equation-based wall distance computation for DES and RANS. Journal of Computational Physics 190(1), 229–248 (2003). https://doi.org/10.1016/S0021-9991(03)00272-9
Van Driest, E.R.: On turbulent flow near a wall. Journal of the Aeronautical Sciences 23(11), 1007–1011 (1956). https://doi.org/10.2514/8.3713
Vassberg, J., and Jameson, A.: Aerodynamic Shape Optimization Part I: Theoretical Background. Introduction to Optimization and Multidisciplinary Design, Von Karman Inst. for Fluid Dynamics Brussels (2006a)
Vassberg, J., and Jameson, A.: Aerodynamic Shape Optimization Part 2: Sample Applications. Introduction to Optimization and Multidisciplinary Design, Von Karman Inst. for Fluid Dynamics Brussels (2006b)
Vreman, A.W.: The adjoint filter operator in large-eddy simulation of turbulent flow. Physics of Fluids 16(6), 2012–2022 (2004). https://doi.org/10.1063/1.1710479
Vreman, A.W.: An Eddy-viscosity subgrid-scale model for turbulent shear flow: algebraic theory and applications. Physics of fluids 16(10), 3670–3681 (2004). https://doi.org/10.1063/1.1785131
Wang, Z.J., Li, Y., Jia, F., Laskowski, G.M., Kopriva, J., Paliath, U., Bhaskaran, R.: Towards industrial large Eddy simulation using the FR/CPR method. Computers & Fluids 156, 579–589 (2017). https://doi.org/10.1016/j.compfluid.2017.04.026
Wilcox, D.C.: Turbulence Modeling for CFD, volume 2. DCW Industries La Canada (1998)
Wukie, N.A., and Orkwis, P.D.: A p-Poisson Wall Distance Approach for Turbulence Modeling. In 23rd AIAA Computational Fluid Dynamics Conference, Denver, Colorado, AIAA 2017-3945. https://doi.org/10.2514/6.2017-3945 (2017)
Yakubov, S., Maquil, T., Rung, T.: Experience using pressure-based CFD methods for Euler–Euler simulations of cavitating flows. Computers & Fluids 111, 91–104 (2015). https://doi.org/10.1016/j.compfluid.2015.01.008
You, D., Moin, P.: A dynamic global-coefficient subgrid-scale eddy-viscosity model for large-eddy simulation in complex geometries. Physics of Fluids. (2007). https://doi.org/10.1063/1.2739419
Zhiyin, Y.: Large-Eddy simulation: past, present and the future. Chinese Journal of Aeronautics 28(1), 11–24 (2015). https://doi.org/10.1016/j.cja.2014.12.007
Acknowledgements
The author would like to gratefully acknowledge the provision of computational resources by Prof. Dr.-Ing. Thomas Rung, Institute for Fluiddynamics and Ship Theory, Hamburg University of Technology.
Funding
Not applicable.
Author information
Authors and Affiliations
Contributions
Not applicable.
Corresponding author
Ethics declarations
Conflict of interest
The author declares that he has no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Ethical Approval
Not applicable.
Informed Consent
Not applicable.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendices
A Standard Laminar Variations
A state perturbation and subsequent isolation is applied to:
-
the continuity equation (1):
$$\begin{aligned} \frac{1}{T} \int _{t^\textrm{s}}^{t^e} \int \hat{p} \frac{\partial \overline{v}_\textrm{k}^\prime }{\partial x_\textrm{k}} \textrm{d} \Omega \textrm{d} t = \frac{1}{T} \int _{t^\textrm{s}}^{t^e} \int \hat{p} \overline{v}_\textrm{k}^\prime n_\textrm{k} \textrm{d} \Gamma \textrm{d} t - \frac{1}{T} \int _{t^\textrm{s}}^{t^e} \int \overline{v}_\textrm{k}^\prime \frac{\partial \hat{p}}{\partial x_\textrm{k}} \textrm{d} \Omega \textrm{d} t \end{aligned}$$(70) -
the momentum equations (2)
$$\begin{aligned} \frac{1}{T} \int _{t^\textrm{s}}^{t^e} \int \hat{v}_\textrm{i} \frac{\partial \rho \overline{v}_\textrm{i}^\prime }{\partial t} \textrm{d} \Omega \textrm{d} t&= \frac{1}{T} \bigg [ \int \hat{v}_\textrm{i} \rho \overline{v}_\textrm{i} \textrm{d} \Omega \bigg ]_{t_s}^{t_e} - \frac{1}{T} \int _{t^\textrm{s}}^{t^e} \int \overline{v}_\textrm{i}^\prime \frac{\partial \rho \hat{v}_\textrm{i}}{\partial t} \textrm{d} \Omega \textrm{d} t \end{aligned}$$(71)$$\begin{aligned} \frac{1}{T} \int _{t^\textrm{s}}^{t^e} \int \hat{v}_\textrm{i} \frac{\partial (\overline{v}_\textrm{k} \rho \overline{v}_\textrm{i})^\prime }{\partial x_\textrm{k}} \textrm{d} \Omega \textrm{d} t&= \frac{1}{T} \int _{t^\textrm{s}}^{t^e} \int \hat{v}_\textrm{i} \rho \left( \overline{v}_\textrm{k}^\prime \overline{v}_\textrm{i} + \overline{v}_\textrm{k} \overline{v}_\textrm{i}^\prime \right) n_\textrm{k} \textrm{d} \Gamma \textrm{d} t \nonumber \\&\qquad - \frac{1}{T} \int _{t^\textrm{s}}^{t^e} \int \overline{v}_\textrm{i}^\prime \left[ \rho \overline{v}_\textrm{k} \frac{\partial \hat{v}_\textrm{k}}{\partial x_\textrm{i}} + \frac{\partial \overline{v}_\textrm{k} \rho \hat{v}_\textrm{i}}{\partial x_\textrm{k}} \right] \textrm{d} \Omega \textrm{d} t \end{aligned}$$(72)$$\begin{aligned} \frac{1}{T} \int _{t^\textrm{s}}^{t^e} \int \hat{v}_\textrm{i} \frac{\partial p^\prime }{\partial x_\textrm{i}} \textrm{d} \Omega&= \frac{1}{T} \int _{t^\textrm{s}}^{t^e} \int \hat{v}_\textrm{i} p^\prime n_\textrm{i} - \int p^\prime \frac{\partial \hat{v}_\textrm{i}}{\partial x_\textrm{i}} \textrm{d} \Omega \textrm{d} t \end{aligned}$$(73)$$\begin{aligned} \frac{1}{T} \int _{t^\textrm{s}}^{t^e} \int \hat{v}_\textrm{i} \frac{\partial }{\partial x_\textrm{k}} \left[ 2 \left( {\mu ^\textrm{t}}^\prime S_\textrm{ik} + \mu ^e S_\textrm{ik}^\prime \right) \right] \textrm{d} \Omega \textrm{d} t&= \frac{1}{T} \int _{t^\textrm{s}}^{t^e} \int \hat{v}_\textrm{i} \left[ 2 \left( {\mu ^\textrm{t}}^\prime S_\textrm{ik} + \mu ^e S_\textrm{ik}^\prime \right) \right] n_\textrm{k} \nonumber \\&\qquad \qquad \qquad - \frac{\partial \hat{v}_\textrm{i}}{\partial x_\textrm{k}} \mu ^e \left( \overline{v}_\textrm{i}^\prime n_\textrm{k} + \overline{v}_\textrm{k}^\prime n_\textrm{i} \right) \textrm{d} \Gamma \textrm{d} t \nonumber \\&\qquad - \frac{1}{T} \int _{t^\textrm{s}}^{t^e} \int {\mu ^\textrm{t}}^\prime 2 S_\textrm{ik} \hat{S}_\textrm{ik} - \overline{v}_\textrm{i}^\prime \frac{\partial }{\partial x_\textrm{k}} \left[ 2 \mu ^e \hat{S}_\textrm{ik} \right] \textrm{d} \Omega \textrm{d} t \end{aligned}$$(74) -
the considered objectives (17):
$$\begin{aligned} {J^f}^\prime&= \frac{1}{T} \int _{t^\textrm{s}}^{t^e} \int \left[ p^\prime \delta _\textrm{ik} - 2 {\mu ^\textrm{t}}^\prime S_\textrm{ik} - 2 \mu ^e S_\textrm{ik}^\prime \right] \textrm{d} \Gamma \textrm{d} t \end{aligned}$$(75)$$\begin{aligned} {J^I}^\prime&= \frac{1}{T} \int _{t^\textrm{s}}^{t^e} \int \overline{v}_\textrm{i}^\prime \left[ \overline{v}_\textrm{i} - \overline{v}_\textrm{i}^\textrm{tar}\right] \textrm{d} \Omega \textrm{d} t \end{aligned}$$(76)
Collecting all (non-objectives) variations yields
B Continuous Adjoint Smagorinsky-Lilly Model
The variation of the Smagorinsky-Lilly subgrid-scale viscosity is splitted in two contributions, viz.
-
the subgrid length scale:
$$\begin{aligned} \frac{1}{T} \int _{t^\textrm{s}}^{t^e} \int \hat{v}_\textrm{i} \frac{\partial }{\partial x_\textrm{k}} \bigg [ 4 \frac{\mu ^\textrm{t}}{l^\textrm{s}} {l^\textrm{s}}^\prime \overline{S}_\textrm{ik} \bigg ] \textrm{d} \Omega \textrm{d} t&= \frac{1}{T} \int _{t^\textrm{s}}^{t^e} \int \hat{v}_\textrm{i} 4 \frac{\mu ^\textrm{t}}{l^\textrm{s}} {l^\textrm{s}}^\prime \overline{S}_\textrm{ik} n_\textrm{k} \textrm{d} \Gamma \textrm{d} t \nonumber \\ {}&\quad - \frac{1}{T} \int _{t^\textrm{s}}^{t^e} \int 4 \frac{\mu ^\textrm{t}}{l^\textrm{s}} {l^\textrm{s}}^\prime \overline{S}_\textrm{ik} \hat{S}_\textrm{ik} \textrm{d} \Omega \textrm{d} t \end{aligned}$$(78) -
the velocity scale, cf. Eq. (33).
Together with the expansion of \({l^\textrm{s}}^\prime\) from Eq. (31), the varied Smagorinsky-Lilly Lagrangian reads
with \(C_{qrik}^a = \overline{S}_\textrm{qr} \overline{S}_\textrm{ik} / (\overline{S}_\textrm{lm} \overline{S}_\textrm{lm})\) and \(C^\textrm{b} = \hat{S}_\textrm{ik} \overline{S}_\textrm{ik} / (\overline{S}_\textrm{lm} \overline{S}_\textrm{lm})\). Assuming \(d^\prime = 0\), \({C^\textrm{s}}^\prime = 0\), \({{l^\textrm{g}}}^\prime = 0\) leads to the following set of continuous adjoint Smagorinsky-Lilly equations
C Further Subgrid Length-Scale Considerations
An improved near wall behavior can be achieved by manipulating the subgrid length scale in line with, e.g., a Van-Driest damping function, i.e.,
where \(y^+\) and \(u^\mathrm {\tau }\) refer to the dimensionless distance to the nearest wall as well as the friction velocity, respectively. The constants are usually assigned to \(A^+ \approx 25\), \(\gamma _1 = 3\), and \(\gamma _2 = 0.5\). Its total variation can be expanded to
where \(t_\textrm{i}\) refers to the flow tangential vector. In line with the primal implementation, the variation of the local length scale couples field quantities with the (dimensionless) nearest wall, which requires additional implementation effort.
D Continuous Adjoint Wall Adapting Local Eddy-Viscosity Model
The variation of the WALE subgrid turbulent viscosity is splitted in two contributions, viz.
-
scaling with the filtered shear rate tensor \(\overline{S}_\textrm{ik}^\prime\), cf. Eq. (49)
-
scaling with its traceless symmetric companion \({\overline{S}_\textrm{ik}^\textrm{d}}^\prime\), cf. Eq. (8), that can be expressed in two ways. Variation of the first version results in six contributions that can be expanded as follows:
Assembling all contributions yields
Further permuting of indices
followed by isolation of \(\overline{S}_\textrm{qr}\) (with \(\alpha _{mi} = \alpha _{im}\)) provides
Adding these –as well as those from Eq. (49)– WALE-related variational contributions to the laminar system (77) yields
with
Assuming the contributions from \({l^\textrm{s}}^\prime\) to vanish leads to the following set of continuous adjoint WALE equations
1.1 D.1 Alternative Derivation
Starting with the second option to express \(S_\textrm{qr}^d\). A perturbation expands to
The calculus is divided into two parts, where the first part \(A_\textrm{qr}\) is further expanded to
Contracting with \(\alpha _\textrm{qr}\) (cf. Eq. (44)) allows for further simplifications, viz.
Raising symmetry arguments (\(\alpha _\textrm{qr} = \alpha _\textrm{qr}\), \(S_\textrm{qr} = S_\textrm{rq}\), \(W_\textrm{qr} = - W_\textrm{rq}\)) finally provides
The second part is considered now, i.e.,
Employing symmetry conditions finally yields
Hence, the required expression \(\alpha _\textrm{qr} {S_\textrm{qr}^d}^\prime\) reads as follows
The condensed variable \(\gamma _\textrm{ik}\) already equals the first part on the r.h.s. of Eq. (107). However, the derivation is continued and (127) plugged into the remaining part of Eq. (47), viz.
Accordingly, the contribution to the adjoint momentum equation reads
and proves the formal correctness of the former derivation, since
Continuous Adjoint Kinetic Energy Subgrid-Scale Model
The variations of the KESS equation (11) are expanded and isolated, viz.
Together with the expansion in (55) and the laminar system Eq. (77), the varied KESS model’s Lagrangian reads
Assuming \({{l^\textrm{g}}}^\prime = 0\) leads to the following set of continuous adjoint KESS equations
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Kühl, N. On the Continuous Adjoint of Prominent Explicit Local Eddy Viscosity-based Large Eddy Simulation Approaches for Incompressible Flows. Flow Turbulence Combust (2024). https://doi.org/10.1007/s10494-024-00543-5
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s10494-024-00543-5