On topological derivatives for elastic solids with uncertain input data (Q1035879)

The authors' approach to the derivation of the worst scenario and the maximum range scenario methods in the present paper generalizes their previous results [\textit{A. A. Novotny, R. A. Feijóo, E. Taroco} and \textit{C. Padra}, Comput. Methods Appl. Mech. Eng. 196, No. 41--44, 4354--4364 (2007; Zbl 1173.74374); \textit{J. Sokołowski} and \textit{A. Żochowski}, ISNM, Int. Ser. Numer. Math. 139, 231--244 (2002; Zbl 1024.49030); \textit{I. Hlaváček, J. Chleboun} and \textit{I. Babuška}, Uncertain input data problems and the worst scenario method. Amsterdam: Elsevier (2004; Zbl 1116.74003)]. Using the concept of topological derivative for the boundary value problems of elasticity in two and three spatial dimensions, the authors apply the topological derivatives to shape and topology optimization within a certain range of input data including the Lamé coefficients and boundary tractions. The derivatives are stable functions, and the concept of topological sensitivity is robust with respect to the imperfections caused by uncertain data. The form of topological derivatives is given for two classes of integral shape functionals: the first for the displacement field, and the second for the stresses with some restrictions on shape functionals in order to assure the existence of topological derivatives. The analytical solutions for stress distribution around a circular (\(N=2\)) and spherial (\(N=3\)) cavities respectively for two and three-dimensional linear elastic bodies are presented in the appendix.