Abstract:The paper presents the thermodynamic response of a composite cylindrical panel on an elastic substrate under thermal shock loading for the first time, owing to the many aeronautical uses of these panels. A novel class of advanced composites known as multi-directional functionally graded materials (MD-FGMs) lacks internal borders and interfacial stress concentrations while having mechanical and structural characteristics that are constantly changing. Because of the exceptional qualities of these cutting-edge materials, this work is the first to investigate the impact of an immediate thermal shock on the thermo-dynamic response of a multi-directional functionally graded (MD-FG) panel lying on an elastic substrate. Peripheral discontinuities at the loaded and unloaded surface boundaries and the shear coupling between the several Winkler spring components are considered in this proposed regime of a Kerr-type elastic substrate. Following the 3D-elasticity technique's extraction of the mathematical governing equations, the equations are solved using the Laplace transform and the three-dimensional differential quadrature method (3D-DQM). Once the equations are solved, the present system's findings are compared to those of a previously published paper, demonstrating a high degree of agreement between the two sets of results. Data-driven solution with the help of the deep neural network (DNN) with input, hidden, and output layers, independent variables, and other related parameters is presented to give a low computational cost method to solve engineering problems like this research. This is done with the help of presented mathematical modeling and verification of the current outputs' results. Lastly, for further study of this structure for aerospace applications, a database including bending information (normal, shear, and displacement fields in different directions) of the MD-FG cylindrical shell under thermal shock loading is provided.

Thermo-dynamic response of multi-directional functionally graded panels resting on an elastic substrate with the aid of data-driven solution