A consistent relaxation of optimal design problems for coupling shape and topological derivatives

In this article, we introduce and analyze a general procedure for approximating a ‘black and white’ shape and topology optimization problem with a density optimization problem, allowing for the presence of ‘grayscale’ regions. Our construction relies on a regularizing operator for smearing the chara...

ver descrição completa

Detalhes bibliográficos
Autores: Amstutz, Samuel, Dapogny, Charles, Ferrer Ferré, Àlex|||0000-0003-1011-0230
Tipo de documento: artigo
Data de publicação:2018
País:España
Recursos:Universitat Politècnica de Catalunya (UPC)
Repositório:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglês
OAI Identifier:oai:upcommons.upc.edu:2117/118037
Acesso em linha:https://hdl.handle.net/2117/118037
https://dx.doi.org/10.1007/s00211-018-0964-4
Access Level:Acceso aberto
Palavra-chave:Structural optimization -- Mathematics
Multiscale modeling--Computer simulation
Level set method
Material interpolation
Shape derivative
Relaxation
Optimal design
topological derivative
Optimització d'estructures
Modelització multiescala
Classificació AMS::74 Mechanics of deformable solids::74P Optimization
Classificació AMS::35 Partial differential equations::35Q Equations of mathematical physics and other areas of application
Classificació AMS::49 Calculus of variations and optimal control
optimization::49Q Manifolds
Classificació AMS::65 Numerical analysis::65N Partial differential equations, boundary value problems
optimization::49M Methods of successive approximations
Àrees temàtiques de la UPC::Matemàtiques i estadística::Topologia
Descrição
Resumo:In this article, we introduce and analyze a general procedure for approximating a ‘black and white’ shape and topology optimization problem with a density optimization problem, allowing for the presence of ‘grayscale’ regions. Our construction relies on a regularizing operator for smearing the characteristic functions involved in the exact optimization problem, and on an interpolation scheme, which endows the intermediate density regions with fictitious material properties. Under mild hypotheses on the smoothing operator and on the interpolation scheme, we prove that the features of the approximate density optimization problem (material properties, objective function, etc.) converge to their exact counterparts as the smoothing parameter vanishes. In particular, the gradient of the approximate objective functional with respect to the density function converges to either the shape or the topological derivative of the exact objective. These results shed new light on the connections between these two different notions of sensitivities for functions of the domain, and they give rise to different numerical algorithms which are illustrated by several experiments