Research On The Application Of Artificial Neural Network In Analysis Of Monitoring Data Of Dams

In recent years, Artificial Neural Network has been developed rapidly. Because of its powerful ability of studying and modifying errors by itself and its characteristic of approaching nonlinear system (in theory), it has been applied to many fields.This paper describes the basic concept of Artificial Neural Network. And it attaches importance to the theory of BP network and its improving algorithm. The paper expatiate the approximation mechanism of Artificial Neural Network and its generalization ability. Based on its characteristic of nonlinear mapping , Artificial Neural Network can be applied to analysis of monitoring data of dams. In the paper, the dam monitoring BP network model is established with the GPS data provided by the GPS Automatic Surveying System to Monitor Geheyan Dam Deformation. The results shows that the predict results of the network model is appreciable superior to those of regression model. Besides, in the paper singular value decomposition is brought forward to resolve the network structure of the dam monitoring BP network model. The example shows that this method is feasible.