Architecture and materials selection in multi-materials design

Abstract

The design process involving both the architecture and the materials represents an hard task mainly due to the high number of potential configurations, thus requiring firstly the development of new and more rigorous approaches but also the development of new tools. To this purpose, we present in this work a new strategy for the design of architectured materials. Such a strategy relies on one hand on the construction of some databases for the selection of both geometrical patterns and materials, and on the other hand on the use of well-known analytical models to describe the physical behaviour of the multi-material. In order to prove its effectiveness, we apply our strategy to the problem of the least-weight design of a multilayer plate that has to meet thermal, electrical and mechanical requirements. Moreover, we use a genetic algorithm, as a numerical tool, to perform the solution search for our problem. Numerical results show that we can obtain optimum configurations characterised by a weight saving up to 59% keeping the same (or even superior) thermal, electrical and stiffness properties than those of a monolithic reference plate.