Study of dynamic response of dams with neural network

Owing to the nonlinear characteristics of dam materials, and the uncertainty of mechanical and physical parameters, it is difficult to describe a real situation of a dam by an ordinary analysis method. Sometimes much efficient information is lost indeed. The purpose of this paper is to modify the aseismatic design method of dams by developing the self-training and self-adjusting characteristics of neural network, using abundance information in the input and output, and advancing the precision of method. In this paper, we introduce a neural network model with recurrent architecture. With the model and the data from the earthquake response of the rock fill dams, we study the feasibility of stimulating the dynamic system with neural network. It is the recurrent component in the architecture that makes the network can describe the dynamic characteristic of the rock fill dams. So the method throw light to the solution of the analysis of the earthquake response of the architecture.