Application of the optimized deep networks in the prediction of the coupled thermoelasticity response of the axially graded composite system subjected to thermal shock loading

Abstract

This research is done to examine the coupled thermoelastic response of composite axially graded beams. Thermal shock is applied axially to the beam. Three-dimensional elasticity theory is used to generate the governing equations of motion. While the Fourier heat conduction relationship and the Lord–Shulman heat transfer theorem are used to construct the time-varying heat conduction connection. Using the discrete singular convolution approach and the Laplace transform, the time-varying equations of motion are solved at design locations. A modified version of the Dubner and Abate method is used to translate the system response from the Laplace domain to the time domain. The Whale Optimization Algorithm and the optimized deep neural network are employed as a prediction mechanism to estimate the system reaction under various scenarios in the final stage. By comparing the outcomes of the procedures utilized with those found in published resources, the correctness of the approaches was confirmed. Results on the doubly thermoelastic response of axially graded composite beams exposed to thermal shock loading are important and are presented in this work. As applicable results for related industries can be observed that by changing the boundary conditions from SS to CC, the location of maximum normal strain tends to be at the center of the axially graded beam