PSI - Issue 52

Made Wiragunarsa et al. / Procedia Structural Integrity 52 (2024) 583–593

590

8

Wiragunarsa et al. / Structural Integrity Procedia 00 (2023) 000–000

(a)

(b)

(d)

(c)

(f)

(e)

(h)

(g)

o , (b) interaction o , (f) interaction

Fig. 6. Crack growth direction on a rectangular plate with a centre crack with various orientation: (a) particle deletion with α = 90

o , (c) particle deletion with α o , (g) particle deletion with α

o , (d) interaction deletion with α o , (h) interaction deletion with α

o , (e) particle deletion with α

deletion with α = 90 deletion with α = 45

= 60

= 60

= 45

o

= 30

= 30

4.2. Stress intensity factor calculation

A quantitative analysis is performed using a rectangular plate with a single edge crack, as shown in Figure 7. The plate is made of aluminium with Young’s modulus E = 73 . 78 GPa , Poisson’s ratio ν = 0 . 33, and density ρ = 2400 kg / m 3 . Dimensions of the plate are L = W = 2 m , and the crack length a is varied. Then, a nominal stress σ = 20 MPa is applied, as shown in the figure. This test analyses the accuracy and convergence of the two modelling techniques. The rectangular plate is discretised using various particle numbers, in which 20x20, 30x30, 40x40, dan 50x50 particles configuration are analysed. The post-processing of the 50x50 particle configuration is presented in Figure 8, in which both modelling techniques yield the same stress distribution. Visually, the particle deletion technique gives a wider crack tip. In order to analyse the state of stress at the crack tip, the stress intensity factor is calculated using the J -integral technique. Then, the normalised stress intensity factor FI = K I σ √ a of the single edge cracked plate using the two crack modelling techniques are presented in Figure 9. The results are compared with the available data in the handbook Murakami [1987]. Based on the test, the interaction deletion technique yields better accuracy and convergence. Two factors are identified that reduced the quality of the particle deletion technique. First, it yields a wider crack tip. The actual crack tip is sharp; when the model is not sharp, the calculated stress state has lower