Structural damage detection and localization using FRF-based model updating approach

This paper discusses the use of a sensitivity-based model updating technique for detecting and locating damage of a structure. Based on global shape correlation (GSC) and global amplitude correlation (GAC) functions, a sensitivity-based updating problem is expressed as a set of linear equations where the coefficient matrix contains the partial derivatives of the correlation functions with respect to the chosen updating parameters. The damage location and severity are identified using a multiple-step procedure based on selecting sensitivity parameters. A simplified railway wheel finite element model with 240 3D solid elements was used for the numerical case study, the damage consisting of modified Young's modulus values for some selected elements. The reference FE model was updated using simulated response functions obtained from the 'damaged' model. A very good response match was obtained between the damaged model and the updated model and it was assumed that the difference between the initial reference model and the damaged, model was indicative of the location and severity of the damage. Several configurations, such as different damage location and severity, were considered. A multi-step procedure was devised to detect the single and multiple damage locations as well as to assess the severity of the damage. The results indicate that the proposed sensitivity-based updating method using incomplete frequency response data can not only detect, but also estimate the amount of structural damage.