PSI - Issue 33

Naoya Oie et al. / Procedia Structural Integrity 33 (2021) 586–597 Oie, N. / Structural Integrity Procedia 00 (2019) 000–000 a 0 / W in Fig. 2, �������� decreases as the plate thickness increases. Comparing the �������� of specimens with the same thickness, �������� decreases with increasing notch depth. The difference between the mean values of �������� for different conditions is not small, approximately a factor of two in the larger case. In other words, the fracture toughness is affected by the scale effect and the notch shape when the fracture toughness is evaluated by quasi CTOD with different three-point bending specimen shapes. As has been pointed out as a disadvantage in the evaluation of fracture mechanics by Minami (2006), the cause of the effect may be due to the large or small plastic constraint on the crack tip. For this reason, only a limited value of specimen geometries, a 0 / W , can be used in the CTOD tests, and the thickness of the specimen and the structure to be evaluated must be the same. 591 6

Fig. 4. Comparison of critical quasi-CTOD for different thicknesses.

3. Local approach In this chapter, the three-point bending test conducted in Chapter 2 is modelled using finite element method (FEM) analysis, and the fracture toughness of the specimen is evaluated using the local approach. The local approach is a method to evaluate the brittle fracture of the entire material based on the local stresses of the material. Therefore, ideally, the fracture toughness should be evaluated using material-specific values independent of the material geometry. 3.1. Finite element analysis The local approach requires information on the local stress field. Nonlinear large deformation elastic-plastic analysis is performed using general-purpose FEM software to obtain information on the local stress and strain fields of the specimen up to fracture. Based on the obtained local distribution information, various evaluation models that have been proposed thus far are applied to the three-point bending test of each steel plate in this study to compare which model can uniformly evaluate the fracture toughness of the material. Due to the material properties required for the 3-D elastic-plastic finite element analysis, the elastic modulus was assumed to be isotropic. The stress-strain relationship was determined by conducting a tensile test at -60°C, the same temperature as the three-point bending test described in Chapter 2. After converting the nominal stress-strain curve into a true stress-true strain curve, the relationship between the plastic strain, which is the true strain minus the elastic strain component, and the true stress was approximated by the n -square hardening rule. Swift's equation shown in equation (7) was used as the approximation equation ( is a constant).