Bayesian updating of tall timber building model using modal data

A framework for the probabilistic finite element model updating based on measured modal data is presented. The described applied to a seven-storey building made of cross-laminated timber panels. experimental estimates forced vibration test are used in process updating. First, generalized Polynomial Chaos surrogate derived representing map from parameters eigenfrequencies and eigenvectors. To overcome difficulties caused by mode switching, we propose novel approach tracking partitioning an extended low-rank representation k shapes resulting different setups into clusters k-means clustering algorithm. Second, with help pairing efficiently perform sensitivity analysis uncertainty quantification first five frequencies corresponding shapes. Finally, surrogate-based Bayesian update performed, providing engineers not only that gives good fit data, but also stochastic represents uncertainties originating initial measuring properties.