Newton's Mesh Independence Principle for a Class Of Optimal Shape Design Problems

DOI10.1137/S0363012996303529MaRDI QIDQ4253024zbMATH OpenFDO

Authors M. Laumen

Publication date 24 June 1999

Published in SIAM Journal on Control and Optimization (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1137/S0363012996303529

zbMATH Keywords

algorithminfinite-dimensional minimization problemslarge scale problemsmesh independence principlenested iterationNewton's methodoptimal shape design

Mathematics Subject Classification ID

Existence theories for optimal control problems involving partial differential equations (49J20) Newton-type methods (49M15) Numerical optimization and variational techniques (65K10)

Review text

Many optimal shape design problems can be stated as infinite-dimensional minimization problems. For deriving an implementable algorithm it has to be decided either to discretize the problem and to use a finite algorithm to solve the discrete problem or to state an algorithm in function space and to discretize this algorithm. However, a fine discretization means that the finite problem consists of many variables, and therefore, an increased amount of work per iteration has to be expected. Keeping the numerical method fixed, the only possibility of reducing the total workload to obtain the discrete solution is to improve the starting point of the nonlinear iteration. Apparently, this can be done by using information from the coarse grid discretization leading to the concept of mesh refinement (respectively, nested iteration). The intention of this paper is to lay the foundation for using mesh refinement to solve the resulting large scale problems efficiently. This is done by proving an appropriately modified mesh independence principle. Shape optimization is described by finding the geometry of a structure which is optimal in the sense of a given minimized cost function with respect to certain constraints. Many-faceted problems naturally arise in engineering applications with the goal of designing a specific structure in an optimal sense, or alternatively, of understanding and determining the shape of a given structure. Typical applications are the design of a nozzle, a thermal diffuser, an airfoil boundary, or various beams and plates, with respect to specific optimality conditions.

Recommendations

• A refined Newton's mesh independence principle for a class of optimal shape design problems
• A comparison of numerical methods for optimal shape design problems
• Newton's Method for a Class of Optimal Shape Design Problems

Cited in

(9)

• On Weak and Strong Convergence of the Projected Gradient Method for Convex Optimization in Real Hilbert Spaces
• Generalized proximal-type methods for weak vector variational inequality problems in Banach spaces
• A strongly convergent method for nonsmooth convex minimization in Hilbert spaces
• A New Moving Mesh Algorithm for the Finite Element Solution of Variational Problems
• A comparison of numerical methods for optimal shape design problems
• Fréchet differentiability of unsteady incompressible Navier-Stokes flow with respect to domain variations of low regularity by using a general analytical framework
• A subgradient-like algorithm for solving vector convex inequalities
• Mesh-Independence of Lagrange-SQP Methods with Lipschitz-Continuous Lagrange Multiplier Updates
• A refined Newton's mesh independence principle for a class of optimal shape design problems

This page was built for publication: Newton's Mesh Independence Principle for a Class Of Optimal Shape Design Problems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4253024)