PSI - Issue 78

Leonardo Casali et al. / Procedia Structural Integrity 78 (2026) 1269–1276

1273

This type of device actually performs in the elastic-plastic range and precisely because of this behavior they had been applied in other projects, as reported in Fulco et. al (2024) and Casali et al. (2023). Indeed in general the device has an hysteretic hardening elastic-plastic behavior as shown in the diagram in Fig. 5. The characteristics of the standardized crescent-shaped nodal dissipative device, corresponding to a simplified bi-linear behavior, are the following: yielding force F y = 200 kN, yielding displacement S y = 8 mm, ultimate displacement S u = 32 mm and corresponding force F u = 255 kN.

250

200

150

100

50

0

-50 Forza (kN)

-100

-150

Maximum displacement Maximum force Equivalent damping

±32 mm ±255 kN

-200

-250

30%

-40

-30

-20

-10

0

10

20

30

40

Spostamento (mm)

Fig. 5 Force-displacement diagram of the standard "crescent moon" device (also referred as "sickle")

In the present project the device is used relying only on its elastic behaviour. The joint connection stiffening is obtained tanks to the behaviour of the end section of the beam that can rotate, at its bottom, around the bearing contact line and is connected at its top to the column by the crescent-shaped device: this arrangement allows to modulate the joint rotational stiffness according to the stiffness of the device and the lever arm with respect to the centre of rotation (arm of the internal forces). The actual stiffness of the system in the elastic field has been determined through pushover analysis on a local model of the beam-column node, evaluating it at the moment corresponding to the yielding point (elastic limit) of the device. This stiffness is dependent on the arm of the internal forces. Configurations of sickles positioned at different heights have been used, therefore there are different values of stiffness and elastic resistant (yielding) moment of the nodal connections. It should be noted that two devices have been positioned for each node to avoid misalignments of the stiffness response. With reference to Block A under consideration, a total number of 163 device is provided distributed as follows. Floors at height 7.00 m: 54 with 0.75 m arm, 61 with 0.90 m arm, 8 with 1.05 m arm. Floors at height 18.50 m: 26 with 0.90 m arm. Floors at height 20.50 m: 14 with 0.90 m arm. Fig. 6 shows views of the numerical model with the positioning of the devices differentiated by type based on colour.

Fig. 6 Positioning of the joint stiffener (represented as spring and differentiated by type based on colour).