PSI - Issue 44

6

Ylenia Di Lallo et al. / Procedia Structural Integrity 44 (2023) 488–495 Y. Di Lallo et al./ Structural Integrity Procedia 00 (2022) 000– 00

493

(a) (b) Fig. 3. Numeric idealization of the façade with the MUDis strategy: (a) discretization of the wall with 250 mm modules (b) view of the software implementation of the model To investigate the in-plane behaviour of the structure, a unidirectional monotonic non-linear static analysis was performed. In the first phase, the vertical loads were statically imposed. In the second phase, an increasing controlled lateral displacement in +X direction was applied both at the top and at the inter-floor level. For the purposes of convergence, the Newton-Raphson method was used, with an energy control criterion for the first phase and a displacement control criterion for the second one. The mechanical properties of the elastic units, summarized in Table 2, are the same as those of the bricks of the original specimen (see Magenes et al. (1995)). Instead, the mechanical parameters of the interface elements, summarised in Table 3, were identified according to the parametric formulas of the modified CI model outlined in Section 2. Table 2. Elastic properties adopted for the units of the MUDis-based model. E G ν γ [N/mm 2 ] [N/mm 2 ] [-] [kg/m 3 ] 16700 7260 0.15 1750

Table 3. Inelastic properties adopted for interface elements of the MUDis-based model. MORTAR JOINT TENSION CUT-OFF COULOMB FRICTION

COMPRESSIVE CAP

k n

k t

σ tra

G

c

φ

G

σ c

C s [-]

G

k p

I

II

III

f

f

f

[N/mm 3 ]

[N/mm 3 ]

[N/mm 2 ]

[N/mm]

[N/mm 2 ]

[°]

[N/mm]

[N/mm 2 ]

[N/mm]

[mm]

16.7

7.2

0.07

0.6

0.1

22.8

0.125

11

9

120

0.09

3.4. Experimental vs. numerical results The first result of this validation process is provided in Fig. 4 where the shear-displacement curve obtained from the numerical analysis described in the previous section is shown. For a better comparison of the results, the backbone experimental curve of the cyclic behaviour provided by Magenes at al. (1995) is juxtaposed with dashed lines on the same graph. As it is possible to observe, the MUDis approach allows reproducing the experimental behaviour in a very satisfactory way and the shear strength of the wall is estimated with good approximation, closely following even the post peak phase. To assess also the model capability of capturing the experimental behaviour in terms of failure mechanisms, Fig. 5