Numerical analysis of functionally graded pressure vessels using stress functions and neural networks

The purpose of this paper is to study the displacements and stresses within radially graded spherical pressure vessels. The body of the sphere is made of a functionally graded material, the material properties are arbitrary functions of the radial coordinate. Our aim is to determine the stress state of the sphere using stress functions while the body is subjected to pressure and prescribed temperature field. The results are used to train a neural network to predict the maximum equivalent stress for the design of the pressure vessel.