PSI - Issue 52

Leonardo Gunawan et al. / Procedia Structural Integrity 52 (2024) 560–569 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

566

7

observed in the experiment, which is in the form of a straight line. The initial damage actually had a straight-line form, but after it reached a certain value, the damage propagation in that direction stopped and an initiation at perpendicular direction took place which resulted in a cross damage form.

Fig. 7. True stress vs true strain of Al 6082 (Yibo et.al, 2013).

Fig. 8. Complete model.

Fig.10 shows the form of failure in the base plate using Model 2 for G C = 400 MPa.mm with varied K. It can be seen that the failure that occurs in the form of a straight line is more closely resembles the result of the experiment. Parametric studies show that the smaller the value of K, or the easier the base panel moves in an up-down direction, the larger the failure size. This can happen because when water pushes the center of the base panel in a direction perpendicular to the panel plane, one of the cross-cut parts, which is on the same side as the side where the bolt is not installed properly, can open wider because that side can move more freely. This indicates that the plates may not be installed symmetrically on all four sides.

Fig. 9. Damage of the base plate Model 1 for (a) =600 (b) =400, (c) =200 (d) =100 MPa.mm. 4.2. Acceleration Fig.11.(a) shows acceleration data from simulations performed using Model 1. The simulation acceleration data truncated at the beginning is caused by smoothing and filtering. The maximum values of acceleration of all models with the varied G C are almost the same. The maximum value is between the maximum values of the simulation and experiment of Francesconi (2009). The trend of the acceleration observed in the simulation is similar to the simulation in reference work but is still somewhat different from the trend in the experiment, namely there is no second peak at t = 7.8 ms.