ABSTRACT
The present proposes a neural network-based method for determining the dynamic characteristic parameters of structures from free vibration data. The method employs the observed dynamic responses to train an appropriate neural network. Conventional back-propagation is used to train the artificial neural network. Then, the modal parameters are directly identified by applying weight matrices in the neural network approach. The modal parameters, including frequencies, damping and vibration shapes of a structure can then be determined accurately. The proposed identification procedure is used to determine the dynamic characteristics of a three-span highway bridge, whose characteristics are determined from its free vibration responses by conducting a field impact test.
