Dynamic behaviour and ANFIS-based natural frequency prediction of honeycomb core sandwich panels for railway coach applications

In this paper, the dynamic analysis of honeycomb-core sandwich plates with carbon and basalt fibre reinforced laminated composite faces, resting on Winkler's and Pasternak's elastic foundations are investigated. The equations of motion are based on Hamilton's principle and solved using the finite element method. Carbon/basalt face layers are predicted by modelling based on the first-order shear deformation theory to predict transverse shears accurately. The reliability of the present numerical model is verified by comparing computed natural frequencies for various vibration modes with published literature data. A comprehensive parametric study has been carried out to investigate the effects of many design parameters such as honeycomb core thickness, honeycomb aspect ratio, cell size, foundation stiffness (Winkler and Pasternak), on the natural frequencies of the sandwich plates for different boundary conditions; clamped–free–free–free, all sides simply supported, two sides simply supported and two sides free, and all sides clamped. Moreover, an adaptive neuro-fuzzy inference system (ANFIS) model is constructed for predicting the first two natural frequency modes of the sandwich plates via some introduced input parameters. Furthermore, an adaptive neuro-fuzzy inference system is developed to predict the first two natural frequency modes of the sandwich plates based on selected input parameters. The proposed ANFIS-based prediction framework offers a reliable and computationally efficient tool for early-stage design optimisation. Overall, the outcomes of this study contribute to the development of lightweight and vibration-efficient honeycomb sandwich structures for Indian railway coach applications, leading to enhanced ride comfort, improved structural integrity, and better energy efficiency.