PSI - Issue 12

M. De Giorgi et al. / Procedia Structural Integrity 12 (2018) 239–248 Author name / Structural Integrity Procedia 00 (2018) 000 – 000

245

7

Since the reliability of the first numerical model has been verified, a larger model was implemented consisting of a 0° laminate having 45x18x8 mm dimensions and three embedded SMA wires. This model was obtained as repetition of a scaled unit cell along the x and y directions. The definition of the material thermal properties was the same for the unite cell except for the heat convection transfer coefficient h that resulted in this case equal to 5.10 W/(m 2 K) according to the formula (1) and considering a more realistic ΔT = 10 K . Also in the case of the composite model with three SMA wires, arising heat power density values were applied: 24x10 6 , 35x10 6 , 45x10 6 , 55x10 6 W/m 3 . Fig. 7 reports the top view of the model for these values of heat input at the end of heating step (t=10s). Moreover, the influence of the heating time was analyzed increasing the heating time up to 40s. The heating curves calculated on the central point of the top surface are reported in Fig. 8 for all the values of power density S.

a)

b)

d)

c)

Fig. 7. Top view of the model with 3 embedded SMAs at time t=10s for power densities S=24·10 6 a), 35·10 6 b), 45·10 6 c), 55·10 6 W/m 3 d).

Fig. 8. Heating curves for all the heat input values on the central point of the top surface.

4. Experimental validation

The reliability of the presented model was verified measuring the thermal response of a GFRP panel embedded with two grids of SMA wires. In detail the panel has dimension of 210 x 300 mm and an overall thickness of about 8 mm. The panel is constituted by 8 plies having a common orientation of 0°, which corresponds to the vertical