Optimisation of anisotropic plates that vary in thicknesses and properties

Jose E. Herencia, Paul M. Weaver, Michael I. Friswell

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

A method to optimise anisotropic composite plates that vary in thickness and properties across the width is presented. The optimisation problem is divided into two levels. At the first level, Mathematical Programming (MP) is applied, where the plate is divided into several strips, each of them modelled using lamination parameters accounting for their anisotropy. The laminate of each of the strips is assumed to be symmetric, or mid-plane symmetric with 0, 90, 45 or -45 degree ply angles. The plate is subjected to a combined loading under strength, buckling and practical design constraints. Manufacturing details of the plate are embedded within the design variables. At the second level, the actual lay-ups of the plate's strips are obtained using a Genetic Algorithm (GA), accounting for manufacturability and design practices. The novelty of this work lies in: the inclusion of anisotropy for elastic tailoring, manufacturing and practical design considerations, as well as the application of the Rayleigh-Ritz (RR) method to assess buckling analysis in plates that vary in thickness and properties across the width.

Original languageEnglish
Title of host publicationProceedings of the 16th International Conference on Composite Materials, ICCM-16 - "A Giant Step Towards Environmental Awareness
Subtitle of host publicationFrom Green Composites to Aerospace"
PublisherInternational Committee on Composite Materials
ISBN (Print)9784931136052
Publication statusPublished - 2007
Externally publishedYes

Publication series

NameICCM International Conferences on Composite Materials

Keywords

  • Anisotropic
  • Buckling
  • Lamination parameters
  • Optimisation
  • Rayleigh-Ritz

Fingerprint

Dive into the research topics of 'Optimisation of anisotropic plates that vary in thicknesses and properties'. Together they form a unique fingerprint.

Cite this