Structural damage identification by a sensitivity-based finite element model updating method using transmissibility function data

This study presents a novel sensitivity-based finite element model updating approach to estimate structural parameters using output-only data. An exact sensitivity relation based on transmissibility data is introduced. Partially measured structural responses are included in mathematical formulations to address incomplete measurement problems and improve the accuracy of the sensitivity relationship. The least-squares algorithm is used to solve the normalised set of equations, and also a weighting approach is used to improve the parameter estimation results. The performance of the proposed method is evaluated based on numerical data of an offshore jacket platform model. In addition to taking the added mass in the calculations, appropriate approaches have been used to select the excitation points and optimal sensor placement. The stability of the proposed method against measurement and mass modelling errors and incompleteness of the measured responses are studied.