Optimization of material composition of nonhomogeneous hollow sphere for thermal stress relaxation making use of neural network

Abstract

In this study, a neural network is applied to optimization problems of material compositions for a nonhomogeneous hollow sphere with arbitrarily distributed and continuously varied material properties such as functionally graded material (FGM). Using the analytical procedure of a laminated hollow sphere model, the analytical temperature solution for the nonhomogeneous hollow sphere is derived approximately. Furthermore, the thermal stress components are formulated under the mechanical condition of being traction free. As a numerical example, the nonhomogeneous hollow sphere composed of zirconium oxide and titanium alloy is considered. Also, as the optimization problem of minimizing the thermal stress distribution, the numerical calculations are carried out making use of neural network, and the optimum material composition is determined taking into account the effect of temperature-dependency of material properties. Furthermore, the results obtained by neural network and ordinary nonlinear programming method are compared.