Structural damage detection using neural network and H∞ filter algorithm

In this paper we propose a neural network-based approach for damage detection of unknown structure systems. Newly developed global H∞ Filter optimal learning algorithm for the neural network to simulate a structural response is developed. This algorithm is based on the worst-case disturbances design criterion, and is therefore robust with respect to model uncertainties and lack of statistical information to the exogenous signals. Simulation results are presented to identify dynamic response characteristics of nonlinear structural systems corresponding to different degrees of parameters changes, which indicate that damage occurred in the structure. It is shown that the proposed method is highly robust and more appropriate in practical early structural damage detection.