PSI - Issue 78

Francesco Nigro et al. / Procedia Structural Integrity 78 (2026) 1537–1544

1542

4. Non-linear static analysis of the RC frame with the exoskeletons The pushover-based assessment of the upgraded structures was performed adopting the same approach already outlined in section 2. Further details about the nonlinear FE models of the steel exoskeletons can be found in Nigro et al. (2023). In the following, the two upgrading scenarios for both the BF and the IF structural models are compared, pointing out the variation of some relevant variables and performance parameters. 4.1. Comparison of performance Firstly, the improvement of the structural performance obtained with the steel exoskeletons was investigated in terms of elastic response parameters. To this end, the pushover curves were transformed into the equivalent bi-linear counterparts, according to the bilinearization procedure suggested by the Italian technical code (D.M. 2018). Besides, the initial stiffness corresponding to the slope of the first linear branch of the bilinear approximation to the pushover curve was transformed into an effective period of vibration T* . Hence, Fig. 4 compares the reduction of the secant-to-yielding period  T* and the increase of the yield acceleration  S ay obtained with the addition of the steel exoskeleton and plotted as percentage of the original values of the RC structure (denoted with the subscript “as - built”) . Fig. 4 shows that steel exoskeletons are able to reduce the secant period at least by 30% for both structures, up to 60% and 45% for BF and IF structures, respectively. The yield value of the effective spectral acceleration is magnified by 100% (as a minimum) for both structures structure up to 700% and 480% for BF and IF structures, respectively . Such variations are always greater for the “X” direction, as the original RC structure is characterised by much lower strength and stiffness in that direction. The lateral force distribution adopted for the pushover analysis does not significantly affect the results. Moreover, the influence of the infill panels is evident, as comparing the BF to the IF structures the variations of period and yield acceleration are lower by factors of 25% and 50%, respectively.

800%

Pushover analysis

Mode X+

Mass X+

Mode Y+

Mass Y+

600%

-70% -60% -50% -40% -30% -20% -10% 0%

400%

 T* / T* as-built

0%  S ay / S ay, as-built 200%

Mode X+

Mass X+

Mode Y+

Mass Y+

Pushover analysis BF - Scenario 1 BF - Scenario 2 IF - Scenario 1 IF - Scenario 2

BF - Scenario 1 BF - Scenario 2 IF - Scenario 1 IF - Scenario 2

a)

b)

Fig. 4. Variation of the secant-to-yielding period T* (a) and the yield acceleration S ay (b) of the SDOF system

Fig. 5 depicts the recorded increments of the safety index for ductile mechanisms for both the DL and SD limit states (Section 2). Fig. 5 highlights that the introduction of the bracing system increases the PGA capacity at least by 50%. Consistently with the observed decrease in the effective vibration period and increase in the yield acceleration, the safety index variations are larger for the X direction at the DL limit state. On the contrary, at the SD limit state the increase of the safety indexes is larger for the Y direction. Differently from observations in terms of elastic response parameters, it can be observed that the influence of infill walls on the ultimate performance is rather small, which could be expected because of the brittle failure of the infill walls. This conclusion holds for the relatively regular structure examined in this study, whereas larger effects of the infill walls might occur in cases of irregular panel distributions.