Structural damage detection by integrating independent component analysis and artificial neural networks

Detection of structural damage is increasingly important for preventing catastrophic failures and prolonging the service life of structures. General approaches for detecting damage include finite-element, principal component analysis, and artificial neural networks (ANN), among others. While these methods have been shown to be effective in certain instances, several unresolved issues hamper their ability to provide reliable damage detection. In this paper, a new approach is presented that integrates independent component analysis (ICA) and ANN to detect changes in structural characteristics. The procedure involves extracting independent components from measured sensor data through ICA and then using these signals as input for a neural network. The experiment presented employs the benchmark data from the University of British Columbia to examine the effectiveness of the method Results indicate the accuracy of damage detection using the proposed method is shown to be far greater than that obtained by the exclusive use of ANN. Furthermore, the prediction output can be used to identify different and levels of damages.