Artificial neural network (ANN) based investigation on the static behaviour of piezo-magneto-thermo-elastic nanocomposite sandwich plate with CNT agglomeration and porosity

Abstract

In this article, a novel finite element method (FEM) – artificial neural network (ANN) approach is adopted to investigate the coupled static parameters (stresses and potentials) of a smart sandwich plate with agglomerated Carbon Nanotubes (CNTs) porous nanocomposite core and piezo-magneto-thermo-electric (PMTE) facings. The main focus of this work is to study the synergistic effects of agglomeration of CNTs, porosity and pyro-coupling of the PMTE materials. Nearly 5184 data points collected from the in-house developed house finite element (FE) computational tool are used to train the ANN model using Levenberg–Marquardt algorithm. The FE formulation is derived using the principle of virtual work in association with higher-order shear deformation theory (HSDT) and von-Karman’s nonlinearity. The externally applied thermal loads are assumed to vary uniformly and linearly across the thickness of the plate. Two states of agglomeration, such as partial and complete; three forms of porosity, such as uniformly distributed, and two variants of functionally graded porosity, have been considered for investigation. Numerical examples are solved to understand the interrelated effects of these material properties on the direct (potentials) and derived (stresses, electric displacement, and magnetic flux densities) static parameters.