Buckling analysis of laminated multiphase composite plates under in-plane loading using data-driven soft computing

Abstract

Laminated multiphase composite plates reinforced with carbon nanotubes (CNTs) and carbon fibers in an epoxy matrix offer excellent mechanical performance for lightweight, high-strength applications. This study focuses on their buckling behavior under in-plane loads by developing a soft computing framework that couples isogeometric analysis (IGA) with Murakami’s zigzag theory for layerwise displacement representation and a deep feedforward neural network for data-driven prediction. A set of numerical simulations is conducted to generate training and validation data across various laminate thicknesses, fiber orientations, CNT/carbon fiber volume fractions, and boundary conditions. These simulation results are split into training and validation subsets, ensuring the evaluation of the machine learning model’s accuracy in predicting critical buckling loads. The trained predictive models are then evaluated against reference solutions to confirm their accuracy. After verifying the models’ reliability, the framework is employed to carry out a detailed parametric investigation, assessing key parameters that affect the buckling response of laminated multiphase composites.