Integrating support vector machine predictions with finite element analysis for linear buckling optimization of composite plates under variable design parameters

The buckling behavior of composite plates plays a critical role in the design of lightweight high strength, high stiffness structures used in aerospace, marine, and civil engineering applications. This study investigates the linear buckling performance of unstiffened and stiffened composite plates made of carbon fiber composite, graphite composite, and Kevlar composite materials. Finite element analysis using ANSYS APDL is employed to evaluate the influence of key design parameters, including fiber orientation, length-to-thickness ratio, aspect ratio, cut-out geometry, stiffener height, stiffener position, stiffener quantity, and stiffener configuration. Results indicate that fiber orientation and laminate configuration significantly affect buckling resistance, with the [0/90/45/−45] layup providing superior performance. Increasing aspect ratio enhances the critical buckling load, whereas larger length-to-thickness ratios reduce structural stability. Circular cut-outs exhibit better buckling performance than rectangular and square cut-outs due to lower stress concentration effects. For stiffened plates, centrally positioned T-shaped stiffeners provide the highest buckling resistance among the configurations studied. Among the investigated materials, Kevlar consistently demonstrates superior buckling performance and lower deformation characteristics. Furthermore, support vector machine (SVM) models are developed to predict critical buckling loads and achieve high prediction accuracy, demonstrating the potential of machine-learning-assisted design optimization. The findings provide useful design guidelines for improving the stability and efficiency of composite plate structures.