Optimal thickness distribution design for blending hybrid composite laminates using Buckling Factor and Failure Index prediction

Abstract

This article introduces an extension to the Optimal Thickness Prediction (OTP) approach for solving the hybrid material composite laminate blending optimization problem considering both Buckling Factor and Failure Index constraints. The proposed optimization approach solves the blending optimization problem with a two-stage procedure. The Stacking Sequence of the laminate is first optimized using optimization algorithms, and then utilized as the input for the CNN-based prediction model to predict the corresponding optimal regional thickness. The present extended approach expands the dimension capacity of the OTP model by including an additional laminate material input dimension and Failure Index output dimension. The additional features broaden the scope of the OTP model to simultaneously handle more design variables and constraints. An integration of the proposed approach with a lamination guideline-based Genetic Algorithm is presented. A hybrid material variation of the 18-panel horseshoe blending optimization problem is introduced and utilized for demonstration of the effectiveness of the proposed approach. The obtained result highlights the significant improvement in performance of the integrated method over the base algorithm.