SOME CIVIL ENGINEERING APPLICATIONS OF NEURAL NETWORKS

This study investigates the feasibility of using neural networks for capturing nonlinear interactions between input and output variables in civil engineering systems. A neural network is an information processing system, the architecture of which essentially mimics the biological system of the brain. The neural network technique is a relatively new computational tool that is particularly useful for evaluating systems with a multitude of non-linear variables. An overview of the neural network methodology is presented and two examples are given. The first example demonstrates the ability of the neural network to learn the known solution to a non-linear engineering relationship; and the other is a practical example involving the shear behaviour of deep reinforced concrete beams. Actual field data were used in training the neural network. The neural network model was found to be as reliable as other conventional methods.