Free vibration behaviour of pomelo peel–inspired sandwich composite plates with additively manufactured cores: an explainable machine learning approach

This research examines the free-vibration behavior of glass-fibre/epoxy sandwich composite plates with a bioinspired polymeric core designed using a pomelo peel structure. It uses a combination of three different methods: experimental, numerical, and data-driven. The face sheet’s elastic properties were experimentally found using the impulse excitation technique, and the shear properties of the PLA core manufactured by additive means were evaluated with an alternate dynamic method. A finite element model based on ANSYS APDL and First-Order Shear Deformation Theory (FSDT) was thoroughly validated against the experimental results for the different boundary conditions with deviations of 6–7%. This validated numerical model was then used to create a large set of parameters

by changing the elastic properties of the materials, shear modulus, density, fiber weight fraction, and laminate configuration. To accurately predict the first-mode natural frequency, regression models were constructed using supervised machine learning on the data collected from the parameters. The Extra Trees Regressor was the most accurate, with an R² value of ~0.997. In addition, Explainable AI methods were employed to demonstrate the importance of each parameter, establishing that the shear modulus, both in-plane and out-of-plane, most significantly impacted the free vibration behavior of the bioinspired sandwich composite plates.

Message: Hi FE, I need help with “Free vibration behaviour of pomelo peel–inspired sandwich composite plates with additively manufactured cores: an explainable machine learning approach” in “Mechanical Engineering”