PSI - Issue 78

Marco Martino Rosso et al. / Procedia Structural Integrity 78 (2026) 301–308

306

block rows to construct the Hankel matrix. Leaving the maximum order as the maximum dimension of the Hankel matrix, the block row parameter was parametrically studied to provide the best resolution of the stabilization diagram and allowing for a reliable modal parameter estimation. The block row parameters were set to 40 for Building 1 and 2 for White Noise 2 and 3, whilst in White Noise 4 it was set to 30 and 35 for Building 1 and 2 respectively, whereas in White Noise 5 it was set to 50 and 30 for Building 1 and 2 respectively, whilst in White Noise 6 it was set to 50 and 35 for Building 1 and 2 respectively, and finally in White Noise 7 it was set to 50 and 40 for Building 1 and 2 respectively. An example of the singular value graph of the power spectral density and the stabilization diagram obtained in white noise 2 in building 1 are illustrated in Figure 3.

Fig. 3. The singular value graph of the power spectral density of FDD and stabilization diagram of SSI-cov algorithm computed with PyOMA software. Label 4 indicates fully stable pole, whilst label 3 indicate stable in frequency and mode shape .

OMA results have been detailed and documented for all the White Noise cases in Tables 3 and 4. In Table 3, the reference modal results from STKO model were reported, and the frequency relative differences were computed. It is worth noting that natural frequency start exhibiting a significant reduction from White Noise 5, thus following Irpinia scaled at 75%. In addition, it is worth reminding that these two buildings were already tested in past, thus they already present cracked elements and experienced accumulated damage. The damage progression was quite evident, since the natural frequency degraded until a reduction of about 17% for the first mode, and even about 52% and 38% for the second and the third mode. The damping ratios estimated with OMA are in line with the expected results of similar structures. [Hz] [Hz] Δ [Hz] [%] [Hz] Δ [Hz] [%] [Hz] Δ [Hz] [%] 1.83 1.91 4.57 5.40 1.91 4.67 3.43 1.78 -2.57 3.55 5.84 6.63 13.59 1.21 6.68 14.45 1.10 6.46 10.66 2.11 8.31 8.98 8.02 5.26 8.96 7.86 5.01 8.53 2.59 4.16 STKO White Noise 5 White Noise 6 White Noise 7 [Hz] [Hz] Δ [Hz] [%] [Hz] Δ [Hz] [%] [Hz] Δ [Hz] [%] 1.83 1.63 -10.91 4.77 1.54 -15.57 2.93 1.51 -17.22 3.65 5.84 5.94 1.81 3.09 5.38 -7.74 4.16 2.76 -52.63 5.66 8.31 7.89 -5.02 3.32 7.15 -13.91 5.66 5.11 -38.48 3.79 Considering Table 4, since Building 2 was equipped with a retrofitting system, it was expected that the structural system would appear stiffer than Building 1. Therefore, the modal parameters do not have a direct reference to the STKO modal frequency results, and, in fact, they were not reported in Table 4. The stiffening effect is evident in the Table 3. OMA experimental results and comparison with STKO reference model for Building 1. STKO White Noise 2 White Noise 3 White Noise 4