PSI - Issue 33

9

Naoya Oie et al. / Procedia Structural Integrity 33 (2021) 586–597 Oie, N. / Structural Integrity Procedia 00 (2019) 000–000

594

���� � � � � � � ��� � ���� � ����

�12� �13� �14�

� � �� � ���� � � � � ���� � � � � � � � �� � ��� � � � � � � � � � � ��� � ��

Fig. 6. The fracture probability of a small reference volume V0: (a) the idea of Beremin; (b) the idea of Bordet.

Fig. 7. Distribution of f(x) .

Fig. 8. SEM observation of the brittle fracture surface.

In this study, we assumed that the initiation point of brittle fracture is limited to the mid-thickness and mid longitudinal direction of the test specimens, and f is a function of the distance from the notch tip, as shown in Fig. 7. The fracture surfaces of 40 specimens were observed by scanning electron microscopy (SEM), and the distance from the notch tip to the initiation point of brittle fracture X was measured, as shown in Fig. 8. The fracture probability that fracture will occur at position P ( x ) was calculated by integrating f near the position of the initiation point. The coefficient k was determined so that the sum of f is 1. The width of dx for integration was set to 30 µm, which is the minimum width of the mesh cutting in FE analysis. 4.2. Validity of the existing Weibull stress models The distribution of P ( X ) for 40 specimens is shown in Fig. 9 for each Weibull stress model. � in Fig. 9 shows the average P ( X ) of each model. It shows that P ( X ) at the actual initiation point is higher for the Bordet and Yoshizu models than for the Beremin model. It is thought that the model that considers crack nucleation by � , such as the