Dynamic monitoring and model updating of a masonry bell tower in Pisa

Since the methods used to identify natural frequencies of existing structures have different resolution capabilities, depending on the algorithms adopted for the dynamic analysis as well as on the type of the dynamic forces used to excite the structure, an appropriate dynamic characterization procedure should combine several signal processing techniques. In the paper, a suitable natural frequencies identification methodology is illustrated in the framework of the dynamic identification of the bell-tower of St. Niccola’s church in Pisa, where classical meth-ods, based on Fourier transform, are combined with more modern methods, based on wavelet transform, integrated by sophisticated finite element investigations as well. The proposed methodology allows to identify the field of application and the soundness of each signal proc-essing technique, achieving an accurate estimation of the main parameters governing the dy-namic structural response, especially when dynamic time histories are characterised by limited amplitude and/or are strongly contaminated by noise.