PSI - Issue 57

Yuki Ono et al. / Procedia Structural Integrity 57 (2024) 290–297 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

296

7

the high-peak load is applied to the surface. The existing numerical approach for HFMI-treated joints is mainly based on simplified smooth geometries [Mikkola et al. 2017, Nazzal et al. 2020, and Ono et al. 2022]; thus, the influence of local material imperfection on residual stress relaxation and on fatigue life estimation has been neglected. In this study, this influence is highlighted by comparing the simplified and actual HFMI geometry models. As shown in Fig. 7, the relaxation behavior of residual stress depends not only on loading histories but also on the definition of geometry model. Applying the high-peak load brings a significant difference between the two models regarding the elastic plastic behavior and the amount of plastic deformation; see Fig. 7. The load sequence effect is revealed more clearly in the actual geometry model. As a result, the large variability in the resulting maximum stress and SWT parameter at a depth of fatigue effective stress can be observed in the actual geometry model. This variability level depends on the size and locations of material imperfection, i.e., level of stress concentration; see Figs. 8 (a) and (c). Compared with the SWT parameter curve as a function of the crack initiation life for HFMI-treated S700 thin plate from Mikkola et al. 2017, all the resulting SWT parameters by the simplified geometry model in this study are below the curve. This implies no fatigue crack initiation despite the fatigue failure happening for loading case 2 which corresponds to a tested condition in Yildirim et al. 2020. Thus, by missing the material imperfection, the model causes significant uncertainties in the local residual stress state and fatigue response, accordingly, cannot estimate the crack initiation and short crack growth periods. Therefore, considering the effect of material imperfection may contribute to developing the robust modelling approach for fatigue life estimation especially for HFMI-treated engineering components and structures that are exposed to high-peak loads. For the maximum stress and fatigue damage below the 100 μm depth or more from the surface, there is a relatively small difference between the two models, as Figs. 8 (b) and (d) indicate. Thus, the simplified geometry model might be utilized for estimating long crack periods after the crack length of 0.1 mm or more, where linear fracture mechanics is assumed to be validated. To highlight the effectiveness of investigated numerical approach in this study, estimating fatigue life and comparison with the fatigue test results needs to be further studied in the future.

1200 1200

400

At 100 μm depth

1000 1000

350

㽢 3.46

㽢 1.00

㽢 1.27

㽢 1.00

800

300

600 600

250

400 400

Location C

200

Case 1 (CAL) Case 2 (1.0 f y → -0.43 f y → CAL) Case 3 (-0.43 f y → 1.0 f y → CAL)

200 200 σ eff,max for actual geometry model (MPa)

150

At depth of effective stress

100 σ max for actual geometry model (MPa) 100

0 0 200 400 600 800 1000 1200 0 200 400 600 800 1000 1200 σ eff,max for simplified geometry model (MPa) 0

100 150 200 250 300 350 400 100 200 200 250 300 350 400 σ max for simplified geometry model (MPa)

Location B

(a) Maximum stress at depth of fatigue effective stress

(b) Maximum stress at 100 μm depth

Case 1 (CAL) Case 2 (1.0 f y → -0.43 f y → CAL) Case 3 (-0.43 f y → 1.0 f y → CAL)

2.5

0.40 0.4

2.50

At 100 μm depth

Location A

2.0

2.00

0.30 0.3

Case 1 (CAL) Case 2 (1.0 f y → -0.43 f y → CAL) Case 3 (-0.43 f y → 1.0 f y → CAL)

㽢 1.32

㽢 1.00

1.5

1.50

㽢 8.85

㽢 1.00

0.20 0.2

1.0

1.00

0.10 0.1

0.50 P SWT for actual geometry model 0.5

P SWT for actual geometry model

At depth of effective stress

0

0.00 0.0

0.00

0.00 0 P SWT for simplified geometry model 0.10 0.1 0.20 0.2 0.30 0.3

0.40 0.4

0.00 0.50 1.00 1.50 2.00 2.50 0.5 1.0 1.5 2.0 2.5 0.0 P SWT for simplified geometry model

(d) SWT parameter at 100 μm depth

(c) SWT parameter at depth of fatigue effective stress Fig. 8 Comparison of maximum stress and SWT parameter between simplified and actual geometry models for Locations A, B, and C