A moving mesh fictitious domain approach for shape optimization problems

A new numerical method based on fictitious domain methods for shape optimization problems governed by the Poisson equation is proposed. The basic idea is to combine the boundary variation technique, in which the mesh is moving during the optimization, and efficient fictitious domain preconditioning in the solution of the (adjoint) state equations. Neumann boundary value problems are solved using an algebraic fictitious domain method. A mixed formulation based on boundary Lagrange multipliers is used for Dirichlet boundary problems and the resulting saddle-point problems are preconditioned with block diagonal fictitious domain preconditioners. Under given assumptions on the meshes, these preconditioners are shown to be optimal with respect to the condition number. The numerical experiments demonstrate the efficiency of the proposed approaches.

[1]  J. Haslinger,et al.  Genetic algorithms and fictitious domain based approaches in shape optimization , 1996 .

[2]  Yuri A. Kuznetsov,et al.  On the solution of the Dirichlet problem for linear elliptic operators by a distributed Lagrande multiplier method , 1998 .

[3]  J. Haslinger,et al.  Finite Element Approximation for Optimal Shape, Material and Topology Design , 1996 .

[4]  Philip E. Gill,et al.  Practical optimization , 1981 .

[5]  James H. Bramble,et al.  The Lagrange multiplier method for Dirichlet’s problem , 1981 .

[6]  G. P. Astrakhantsev.  Method of fictitious domains for a second-order elliptic equation with natural boundary conditions , 1978 .

[7]  Jacques Periaux,et al.  On some imbedding methods applied to fluid dynamics and elecjro-magnetics , 1991 .

[8]  Christoph Börgers,et al.  A triangulation algorithm for fast elliptic solvers based on domain imbedding , 1990 .

[9]  Anders Klarbring,et al.  Fictitious domain/mixed finite element approach for a class of optimal shape design problems , 1995 .

[10]  I. Babuska.  The finite element method with Lagrangian multipliers , 1973 .

[11]  M. Saunders,et al.  Solution of Sparse Indefinite Systems of Linear Equations , 1975 .

[12]  Olivier Pironneau,et al.  Fictitious domains with separable preconditioners versus unstructured adapted meshes , 1992, IMPACT Comput. Sci. Eng..

[13]  V. Komkov.  Optimal shape design for elliptic systems , 1986 .

[14]  Claude Fleury,et al.  Sensitivity Analysis With Unstructured Free Mesh Generators in 2-D Shape Optimization , 1993 .

[15]  J. Zolésio,et al.  Introduction to shape optimization : shape sensitivity analysis , 1992 .

[16]  Yuri A. Kuznetsov,et al.  Fictitious Domain Methods for the Numerical Solution of Two-Dimensional Scattering Problems , 1998 .

[17]  R. Glowinski,et al.  A fictitious domain method for Dirichlet problem and applications , 1994 .

[18]  Tuomo Rossi,et al.  A Parallel Fast Direct Solver for Block Tridiagonal Systems with Separable Matrices of Arbitrary Dimension , 1999, SIAM J. Sci. Comput..

[19]  Raino A. E. Mäkinen,et al.  Finite-element design sensitivity analysis for non-linear potential problems , 1990 .

[20]  Panayot S. Vassilevski,et al.  Preconditioning Capacitance Matrix Problems in Domain Imbedding , 1994, SIAM J. Sci. Comput..

[21]  R. A. Brockman,et al.  Geometric sensitivity analysis with isoparametric finite elements , 1987 .

[22]  Jan Sokolowski,et al.  Introduction to shape optimization , 1992 .

[23]  P. Swarztrauber.  THE METHODS OF CYCLIC REDUCTION, FOURIER ANALYSIS AND THE FACR ALGORITHM FOR THE DISCRETE SOLUTION OF POISSON'S EQUATION ON A RECTANGLE* , 1977 .

[24]  Jaroslav Haslinger.  Fictitious Domain Approaches in Shape Optimization , 1998 .

[25]  Michal Kočvara,et al.  Control/fictitious domain method for solving optimal shape design problems , 1993 .

[26]  Jiwen He,et al.  Méthodes de domaines fictifs en mécanique des fluides : Applications aux écoulements potentiels instationnaires autour d'obstacles mobiles , 1994 .

[27]  O. Widlund,et al.  On finite element domain imbedding methods , 1990 .

[28]  T. Chan.  Analysis of preconditioners for domain decomposition , 1987 .

[29]  G. Marchuk,et al.  Fictitious domain and domain decomposition methods , 1986 .

[30]  J. Haslinger,et al.  Numerical realization of a fictitious domain approach used in shape optimization. Part I: Distributed controls , 1996 .

[31]  YU. A. KUZNETSOV,et al.  Efficient iterative solvers for elliptic finite element problems on nonmatching grids , 1995 .

[32]  V. Braibant,et al.  Shape optimal design using B-splines , 1984 .

[33]  Jacques Periaux,et al.  Numerical simulation and optimal shape for viscous flow by a fictitious domain method , 1995 .

[34]  Jaroslav Haslinger,et al.  Imbedding/Control Approach for Solving Optimal Shape Design Problems , 1993 .

[35]  J. Pasciak,et al.  The Construction of Preconditioners for Elliptic Problems by Substructuring. , 2010 .

[36]  Anne Greenbaum,et al.  Iterative methods for solving linear systems , 1997, Frontiers in applied mathematics.

[37]  Laurent Tomas.  Optimisation de forme et domaines fictifs : Analyse de nouvelles formulations et aspects algorithmiques , 1997 .