PSI - Issue 28

Saiaf Bin Rayhan et al. / Procedia Structural Integrity 28 (2020) 1892–1900 Author name / Structural Integrity Procedia 00 (2019) 000–000

1896

5

boundary conditions can also be employed to calculate the mechanical properties. Finally, the RVE is exposed to several load cases and the responses are computed accordingly to obtain homogenized material data. A. B. C.

Fig. 1. RVE geometries with block meshing A: Square shape; B: Diamond shape; C: Hexagonal shape.

5. Result analysis and discussion 5.1. Longitudinal Young’s Modulus, E 11

At first, we will examine the longitudinal Young’s Modulus (E 11 ) data for different fiber volume fraction, obtained from Ansys and compare them with available analytical, experimental and finite element outcomes (Younes et al. (2012)). It is evident that, for both Carbon-Epoxy, Fig. 2 (A) and Polyethylene-Epoxy, Fig. 2 (B), Ansys Material Designer’s prediction agrees quite well with other outcomes and the estimated values are identical for all RVE shapes.

65

250

A.

B.

52

200

39

150

Exp. Chamis/Bridging EAM M-T S-C

E 11 ,GPa

Exp. Chamis/Bridging EAM M-T S-C

E 11 ,GPa

26

100

13

50

Comsol FE. (Sq./Dm./Hex.) Ansys FE. (Sq./Dm./Hex.)

Comsol FE. (Sq./Dm./Hex.) Ansys FE. (Sq./Dm./Hex.)

0

0

0

0 , 2

0 , 4

0 , 6

0 , 8

1

0

0 , 2

0 , 4

0 , 6

0 , 8

1

Fiber Volume Fraction,V f

Fiber Volume Fraction, V f

Figure 2. A: Longitudinal Young’s Modulus (E 11 ), Carbon-Epoxy; B: Longitudinal Young’s Modulus (E 11 ), Polyethylene-Epoxy

5.2. Transversal Young’s Modulus, E 22 Unlike the previous case study, for transversal Young’s Modulus, E 22 , disagreements are found among analytical, experimental and numerical predictions, Fig. 3 (A) and 3 (B). More interestingly, RVE shapes of Material Designer predict slightly different outcomes (maximum difference found 5.14%) for the Carbon-Epoxy case, Fig. 3 (A). However, this is not true for Polyethylene-Epoxy where the results are identical, Fig. 3 (B). For the Carbon-Epoxy case, among the three different RVE shapes, square-shaped RVE of Material Designer provides excellent agreement with experimental data with a maximum difference of 1.85% found for 0.65 fiber volume fraction. For Polyethylene-Epoxy, in general, Material Designer estimates better results than Comsol FE and other analytical solutions in comparison to experiments, except the Chamis model, which shows slightly better predictions for this specific case. However, the highest discrepancy of estimated data by Material Designer with the experiment is 5.85% for 0.68 fiber volume fraction.