基于BP神经网络组合套筒的粘弹性分析

he viscoelastic material has a strong nonlinear, and its differential constitutive equation has been displayed based on the generalized Kelvin model.According to the boundary conditions of the viscoelastic sleeve,this paper deduces the Laplace equation of viscoelastic response in the plane strain.Training the results of orthogonal design test by three layer BP neural network by an example can safely draw a conclusion that the trained network can easily realize the mapping of viscoelastic material test set.It shows that the results of neural network test and numerical solution are almost equal with a maximum error less than 4.5 percent. This paper illustrate that the trained neural network can effectively solve the nonlinear problem of viscoelastic material.