PSI - Issue 44

Samuel Barattucci et al. / Procedia Structural Integrity 44 (2023) 426–433 Barattucci et al/ Structural Integrity Procedia 00 (2022) 000–000

431

6

(a) (b) Fig. 4. Observed failure mode: (a) for a level of drift = 0.5%; (b) for a level of drift = 2%.

shear cracking strength F cr . The second branch of the envelope runs from the first cracking point up to the point of maximum strength F max , assumed equal to 1.30 F cr , and it is characterized by the secant stiffness K sec . The third branch of the envelope is the post-capping degrading branch, which runs from the maximum strength to the residual strength. The stiffness K deg of this branch is assumed equal to a rate α of K el and the residual strength F res is determined as a rate β of F max . Finally, the fourth horizontal branch reproduces the infill residual strength. The abovementioned multi-linear force-displacement relationship is then converted into an equivalent stress-strain relationship. The values of stress and strain corresponding to the three corners of the envelope both in the positive and negative direction are assigned to the truss member by means of the “Pinching4” uniaxial material. The cyclic response under horizontal force of the numerical model was determined by means of a cyclic nonlinear static analysis run under displacement control. The top displacement was applied according to the loading protocol used in the experimental test. 4.2. Proposed technique for calibration of the numerical model The mechanical features of the numerical model previously described were calibrated so that the predicted response matched that provided by the experimental test. The results of the experimental test, in terms of base shear and top displacement, were assumed as target. Based on the observation of the experimental results, the calibration of the numerical model was conducted following a “sub-assembly” approach. Since the gradual reduction of stiffness and strength of infill panels is generally observed in almost all r.c. framed structures, the proposed calibration approach could be widely adopted. As first step, the parameters characterizing the r.c. columns and beam were determined assuming as

σ

σ

ε

ε

(a)

(b) (c) Fig. 5. Numerical model: (a) Concrete 01, (b) Steel 02, (c) monotonic response of infill panel