Efficient machine-learning algorithm applied to predict the transient shock reaction of the elastic structure partially rested on the viscoelastic substrate

Abstract

Due to numerous applications of piezoelectric materials in the modern technologies, this study assesses the thermomechanical shock behavior of the functionally graded graphene platelets reinforced nanocomposite (FG-GPLRN) annular plate surrounded by two piezoelectric layers and partially rested on the viscoelastic substrate for different cases of boundary conditions by metaheuristic optimized machine-learning methods, for the first time. Thermal and mechanical shocks are simultaneously applied on the upper surface of the mentioned structure. Governing equations of the system are formulated in the background of three-dimensional (3D) elasticity theory. Energy balance of the system is considered based on the Lord-Shulman theorem. Differential quadrature method (DQM) is selected as the main solver to determine the spatial response of the system from the state-space form of the governing differential equations. Additionally, Laplace transform is collaborated with modified Dubner and Abate’s approach to predict the temporal response of the system. Accuracy of the applied methods is carefully examined and verified through comparative study performed between the current results and those determined in the published high-quality studies. Valuable outcomes of the current approach would be directly employed in designing process of similar structures interacting with probable thermomechanical shocks.