Dynamic identification and model updating of an eight-storey CLT building

•The results of the dynamic identification of an eight-storey CLT building are reported and interpreted in the light of a simplified analytical model.•The analytical model is calibrated on the experimental modal parameters using a proper objective function, whose estimand parameters were the masses of the storeys.•The connections have a negligible influence on the building dynamics in operational conditions.•The CLT floors behave like a rigid diaphragm.•The damping factors of the building under test are very close to the damping of the timber itself, which is about 2%. So far, few in situtests have been carried out to estimate the modal parameters of multi-storey Cross-Laminated-Timber (CLT) buildings: the understanding of their dynamic behaviour under operational conditions is still an open issue. In this paper, the results of the dynamic identification of an eight-storey CLT building are reported and interpreted in the light of a simplified shear-type analytical model. The structural model is calibrated on the experimental modal parameters, which are assessed using the Stochastic Subspace Identification (SSI). The SSI provides mode shapes, natural frequencies and damping factors of the structures in its first linear response range; The minimum of a modal-based objective function gives an estimation of the unknown parameters of the structural model. Given the results of the dynamic identification, the storey-masses are the chosen optimization parameters: the values of the experimental eigenfrequencies likely reveal a continuum-like behaviour of the building without connections. The identified storey masses are indeed realistic and confirm the negligible role of the connections in the low-amplitude dynamic response. The nonlinear effects of connections over the building response to higher displacements are discussed, providing simplified and practice-oriented correlations for the first natural frequency estimation.