Two Approaches to Structural Damage Identification: Model Updating versus Soft Computing

Abstract

This article presents two approaches for structural damage identification, each based on a different philosophy. The virtual distortion method (VDM) is a model-updating method of damage assessment, utilizing gradient-based optimization techniques to solve the resulting inverse dynamic problem in the time domain. Case-based reasoning (CBR) is a softcomputing method utilizing wavelet transformation for signal processing and neural networks for training a base of damage cases to use for retrieving a similar relevant case. Advantages and drawbacks of each approach are discussed. Successful calibration of a numerical model from experiments has been shown as a sin equa non for the VDM approach. A numerical example of a beam is presented including a demonstration of the complexity of the inverse problem. Qualitative and quantitative comparisons between the two approaches are made.