Nonlinear identification of dynamic systems using neural networks

A neural-network-based method is proposed for the modeling and identificati on of a discrete-time nonlinear hysteretic system during strong earthquake motion. The learning or modeling capability of multilayer neural networks i s explained from the mathematical point of view. The main idea of the propo sed neural approach is explained, and it is shown that a multilayer neural network is a general type of NARMAX model and is suitable for the extreme n onlinear input-output mapping problems. Numerical simulation of a three-sto ry building and a real structure (a bridge in Taiwan) subjected to several recorded earthquakes are used here to demonstrate the proposed method. The results illustrate that the neural network approach is a reliable and feasi ble method.