پيش بيني هوشمند فراآب ناشي از پايه پل با استفاده از شبكه عصبي مصنوعي مبتني بر تابع پايه شعاعي

Intelligent Prediction of Afflux Due to Bridge Pier Using Artificial Neural Networks Based on Radial Basis Functions

Depth of water at the upstream of pier will increase when a bridge is constructed. The difference between water depth before and after the bridge construction is usully termed as afflux. The purpose of this paper is to apply and evaluate the abilities of Artificial Neural Networks (ANN) to predict this phenomena. Mapping between input data (discharge,normal depth,length to width ratio of bridge pier,contractin ratio and angle of pier axis) and output data(afflux) has been provided by a Radial Basis Function(RBF) ANN based on imperical data produced by experimental measurements. The results show that a RBF artifical neural network including two hidden layers can predict intelligently the afflux and its performance is much better than other conventional approches.