Adjoint volume-of-fluid approaches for the hydrodynamic optimisation of ships

ABSTRACT The paper is concerned with simulation-based shape optimisation in marine engineering flows. Attention is devoted to the derivation of an adjoint complement to two-phase flow finite-volume procedures. The strategy refers to an extension of a hybrid continuous/discrete adjoint method for volume-of-fluid (VoF) approaches. The study outlines means to formulate a discrete adjoint VoF scheme from the terms that originate from variations of the fluid properties. The adjoint solution is verified against results of a direct differentiation technique. The application is devoted to the drag optimisation of the Kriso container ship. A kernel-based self-parametrisation approach of the design surface is combined with mesh-morphing techniques to drive the 150,000 shape parameters of the vessel without the need to revisit or differentiate the CAD environment during the optimisation. The optimisation process obeys to practical constraints. The optimised vessel displays more than 5% reduction of total drag while maintaining main dimensions and displacement.

[1]  D. H. Kim,et al.  Measurement of flows around modern commercial ship models , 2001 .

[2]  Adrian Sandu,et al.  Analysis of Discrete Adjoints for Upwind Numerical Schemes , 2005, International Conference on Computational Science.

[3]  Arthur Stück,et al.  Adjoint complement to viscous finite-volume pressure-correction methods , 2013, J. Comput. Phys..

[4]  B. P. Leonard,et al.  The ULTIMATE conservative difference scheme applied to unsteady one-dimensional advection , 1991 .

[5]  Nikolaus A. Adams,et al.  Anti-diffusion method for interface steepening in two-phase incompressible flow , 2011, J. Comput. Phys..

[6]  Juan J. Alonso,et al.  Shape Sensitivity of Free-Surface Interfaces Using a Level Set Methodology , 2012 .

[7]  O. Ubbink.  Numerical prediction of two fluid systems with sharp interfaces , 1997 .

[8]  Julia Springer.  Multiphase adjoint optimization for efficient calculation of rigid body positions in Navier-Stokes flow , 2014 .

[9]  Tadeusz Koronowicz,et al.  COMPARISON OF CICSAM AND HRIC HIGH-RESOLUTION SCHEMES FOR INTERFACE CAPTURING , 2008 .

[10]  C. Othmer.  A continuous adjoint formulation for the computation of topological and surface sensitivities of ducted flows , 2008 .

[11]  Sophie Papst,et al.  Computational Methods For Fluid Dynamics , 2016 .

[12]  C. W. Hirt,et al.  Volume of fluid (VOF) method for the dynamics of free boundaries , 1981 .

[13]  Antony Jameson,et al.  Aerodynamic design via control theory , 1988, J. Sci. Comput..

[14]  T. Rung,et al.  Hybrid MPI/OpenMP parallelization of an Euler–Lagrange approach to cavitation modelling , 2013 .

[15]  Joaquim R. R. A. Martins,et al.  The complex-step derivative approximation , 2003, TOMS.

[16]  J. Periaux,et al.  Evolutionary Methods for Design, Optimization and Control with Applications to Industrial Problems , 2001 .

[17]  Kyriakos C. Giannakoglou,et al.  The continuous adjoint approach to the k–ε turbulence model for shape optimization and optimal active control of turbulent flows , 2015 .

[18]  H. Rusche.  Computational fluid dynamics of dispersed two-phase flows at high phase fractions , 2003 .

[19]  R. Dwight,et al.  Numerical sensitivity analysis for aerodynamic optimization: A survey of approaches , 2010 .

[20]  LUIGI MARTINELLI,et al.  An adjoint method for design optimization of ship hulls , 2007 .

[21]  B. V. Leer,et al.  Towards the ultimate conservative difference scheme V. A second-order sequel to Godunov's method , 1979 .

[22]  Rainald Löhner,et al.  An adjoint‐based design methodology for CFD problems , 2004 .

[23]  Lars Larsson,et al.  Numerical Ship Hydrodynamics - An Assessment of the Gothenburg 2010 Workshop , 2014 .

[24]  S. Nadarajah,et al.  The discrete adjoint approach to aerodynamic shape optimization , 2003 .

[25]  Thomas Rung,et al.  Sub-cycling Strategies for Maritime Two-Phase Flow Simulations , 2013 .

[26]  O. Pironneau.  On optimum design in fluid mechanics , 1974 .

[27]  Christian Kahle,et al.  A Nonlinear Model Predictive Concept for Control of Two-Phase Flows Governed by the Cahn-Hilliard Navier-Stokes System , 2011, System Modelling and Optimization.

[28]  Harald Garcke,et al.  A stable and linear time discretization for a thermodynamically consistent model for two-phase incompressible flow , 2014, 1402.6524.

[29]  Kyriakos C. Giannakoglou,et al.  Continuous adjoint approach to the Spalart–Allmaras turbulence model for incompressible flows , 2009 .

[30]  Juan J. Alonso,et al.  Design of free-surface interfaces using RANS equations , 2013 .

[31]  F. Menter,et al.  Ten Years of Industrial Experience with the SST Turbulence Model , 2003 .

[32]  Cecil Armstrong,et al.  Optimizing parameterized CAD geometries using sensitivities based on adjoint functions , 2012 .

[33]  A. Jameson.  Optimum aerodynamic design using CFD and control theory , 1995 .

[34]  Arthur Stück,et al.  Adjoint‐based design of shock mitigation devices , 2009 .

[35]  Arthur Stück,et al.  Adjoint Navier-Stokes methods for hydrodynamic shape optimisation , 2012 .

[36]  石井 昌憲.  NASA Langley Research Center 滞在記：バージニア州車事情 , 2014 .

[37]  Thomas Rung,et al.  Computation of mechanically coupled bodies in a seaway , 2017 .

[38]  Kai-Uwe Bletzinger,et al.  A consistent frame for sensitivity filtering and the vertex assigned morphing of optimal shape , 2014, Structural and Multidisciplinary Optimization.

[39]  T. Rung,et al.  CAD-free hydrodynamic optimisation using consistent kernel-based sensitivity filtering , 2015 .

[40]  Arthur Stück,et al.  Adjoint-based Hull Design for Wake Optimisation , 2011 .

[41]  Carsten Othmer,et al.  Adjoint methods for car aerodynamics , 2014, Journal of Mathematics in Industry.

[42]  Sergey Yakubov,et al.  Experience using pressure-based CFD methods for Euler–Euler simulations of cavitating flows , 2015 .

[43]  Thomas Rung,et al.  Analysis of non-conservative interpolation techniques in overset grid finite-volume methods , 2017 .

[44]  T. Rung,et al.  Adjoint RANS with filtered shape derivatives for hydrodynamic optimisation , 2011 .

[45]  Cecil Armstrong,et al.  Linking Adjoint Sensitivity Maps with CAD Parameters , 2008 .