PSI - Issue 64

Vera Rillo et al. / Procedia Structural Integrity 64 (2024) 700–707 Rillo et al./ Structural Integrity Procedia 00 (2019) 000 – 000

705

6

one sensor is deactivated every time, the modal shapes are reconstructed interpolating the missing data. In particular, a spline interpolation was used as curve-fitting method implemented in MATLAB (2022). It was observed that, despite deactivating various sensors, the frequencies did not undergo significant variations compared to the initial configuration with 24 active sensors. Furthermore, from the values along the diagonal of the MAC, it was observed that the higher modes (from the forth to the sixth one) were poorly estimated when the number of sensors is reduced. This is also due to the difficulty in the interpolation due to the large dispersion assumed by the points. In general, it was found that, to best estimate at least the first 3 modes it is necessary to install 13 using the ADPR ranking. Moreover, since the authors have only 10 prototype sensors to be located on the bridge under study for a further monitoring, it was decided to investigate the quality of the modal shapes obtained by activating only the first 10 sensors for each ranking performed. Therefore, the MAC for each configuration with only 10 sensors was calculated for a comparison. Mode 1 f=1.807 Hz Mode 2 f=2.734 Hz

Mode 3 f=2.93Hz

Mode 4 f=3.516 Hz

Mode 5 f=4.614 Hz

Mode 6 f=5.469 Hz

Fig. 3. Experimental simulated modes

In Fig. 4 the best configurations with 10 sensors from the ADPR, EFI and EVP ranking, respectively, are shown.