PSI - Issue 23

Atri Nath et al. / Procedia Structural Integrity 23 (2019) 263–268 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

267 5

Kourosis, 2013

Present work

 a =530 MPa

 a =550 MPa

 m =70 MPa

 -control

0

1

2

3

4

5

F error %

TA16 alloy

(a)

(c)

(d)

(b)

Fig. 3: Application of the suggested methodology to TA16 titanium alloy for a) monotonic b) strain-controlled c) ratcheting d) accuracy estimation using F error

Agius et al. (2017) used a modified multicomponent Armstrong-Frederick model with multiplier (mMAFM) for predicting the cyclic-plastic response of AA7075 aluminum alloy considering of a lower yield stress (435 MPa) than estimated from the tensile half of the first cycle (500 MPa). This study uses the yield value for monotonic response reported by Agius et al. (2017) (i.e. 500 MPa) for cyclic-plastic predictions. The suggested methodology provides improved ratcheting predictions for AA7075 as observed in Fig. 4b, along with comparable simulated strain-controlled hysteresis loop simulations (Fig. 4a). The obtained F error for the cyclic-plastic simulations (summarized in Fig. 4c) clearly demonstrate the superior predictive capability of the current methodology over the one reported by Agius et al. (2017). The magnitude of F error computed for the simulated stabilized hysteresis loop obtained by using the present methodology is 2% in comparison with 5% obtained for the simulations reported by Agius et al. (2017). Similarly, the maximum F error for the ratcheting predictions by Agius et al. (2017) is 1.9% compared to only 1.6% by the current methodology.

Agius et al., 2017

Present work

 a =50 MPa

 a =40 MPa

 m = 500MPa

 -control

0 1 2 3 4 5 F error %

AA7075-T6 alloy

(a) (c) Fig. 4: Application of the suggested methodology to AA7075 T6 aluminum alloy for a) strain-controlled b) ratcheting c) accuracy estimation using F error (b)

4. Discussion

The previous section qualitatively and quantitatively compared the experimentally observed cyclic-plastic response of CS1026 steel, SA333 steel, AA7075 aluminum alloy, and TA16 titanium alloy with the predictions using the proposed methodology as well as other reported numerical ones (Agius et al., 2017; Bari and Hassan, 2000; Hassan and Kyriakides, 1992; Kan et al., 2011; Kourousis, 2013; Paul et al., 2010) . A single set of CIKH model parameters were used for prediction of both monotonic and cyclic-plastic deformation for the investigated materials. However, a marked difference in the contribution of the backstresses to plastic response for ferrous and non-ferrous materials is noted as summarized in Fig. 5. The contribution of the fourth backstress component of the kinematic hardening is larger for non-ferrous materials than that for ferrous materials. From the CIKH parameters summarized in Table 2, it is also observed that the threshold term related to the 4 th backstress ( α 4 ) is of higher magnitude for non-ferrous