Multiobjective stacking sequence optimization for unsymmetric laminated plates under several empirical constraints

Abstract

This study proposes a multiobjective stacking sequence optimization method for an unsymmetric laminated plate using for a torsion-box structure of an aircraft wing. The objective functions are to minimize the thermal warping occurred during molding process and the bending stiffness ratio that empirically corresponds to the compression after impact strength. The constraints considered are a limited set of ply orientation angles, the allowable number of contiguous plies, and the allowable orientation angle difference between neighboring plies that empirically derived from several strength criteria as delamination, matrix crack and so on. Thus, the design problem is formulated as a combinatorial multiobjective optimization problem with several empirical constraints. To solve this problem, this study adopts a multiobjective genetic algorithm. To consider the empirical constraints on the genetic algorithm, new searching strategies that arrange mutation and gene repair strategy are proposed. Furthermore, lamination parameters and binary trees are introduced to improve searching performance. Through numerical examples, the performance and efficiency of the proposed method are demonstrated. Then, features of the obtained Pareto set are discussed for several design conditions with different thicknesses.

Keywords

• unsymmetric laminated plates
• empirical constraints
• evolutionary multiobjective optimization method
• lamination parameters
• repair strategy
• adaptive mutation
• binary tree