MINIMUM THICKNESS DESIGN OF COMPOSITE LAMINATES SUBJECT TO BUCKLING AND STRENGTH CONSTRAINTS BY GENETIC ALGORITHMS

摘要

In an earlier paper, the authors have shown that using a binary tree for storing past designs and employing approximation-based local improvement substantially reduces the cost of genetic algorithms for buckling load maximization of composite plates. This paper describes the application of the same techniques to the stacking sequence optimization of a laminated composite plate for thickness minimization. The thickness minimization requires use of variable length genetic strings with an objective function that includes both the thickness and constraint margins. The design variables are coded orientations in the lamination sequence. Genetic algorithms are shown to be effective for producing several alternative designs. The binary tree and local improvement techniques are shown to help in substantially reducing the number of required analyses.