PSI - Issue 21

Hande Yavuz / Procedia Structural Integrity 21 (2019) 112–119 H. Yavuz/Structural Integrity Procedia 00 (2019) 000 – 000

115

4

composites under prescribed boundary conditions.

10000 constraint 1 dominant

Metals and alloys

1000

Beryllium, grade I-250, hot isostatically pressed

100

Epoxy/E-glass fiber, UD prepreg, UD lay-up

Beryllium (50-127 micron, f)

Epoxy/aramid fiber, UD prepreg, UD lay-up BMI/HS carbon fiber, UD prepreg, UD lay-up

Composites

BMI/HS carbon fiber, UD prepreg, UD lay-up

Epoxy/HS carbon fiber, UD prepreg, UD lay-up

Cyanate ester/HM carbon fiber, UD prepreg, UD lay-up PEEK/IM carbon fiber, UD prepreg, UD lay-up

10

constraint 2 dominant

100 M1=Density / Yield strength (elastic limit)^(1 / 2)

1000

10000

M2=Density / Young's modulus^(1 / 3)

Fig. 1. Coupling line and best choice of materials for skin panels using multiple constraints design.

2.2. Computational damage analysis for composite laminates The damage analysis of various unnotched composite laminates were explored by both MATLAB and Abaqus/Standard v6.14. The built-in Hashin model in Abaqus was chosen to evaluate the ply-level mechanical damage initiation characteristics of various notched composite laminates. According to Hashin’s theory, fo ur different damage initiation mechanisms were considered: HSNFTCRT, HSNFCCRT, HSNMTCRT, and HSNMCCRT (Hashin 1980). Without damage evolution model, Abaqus simply keeps track of the damage initiation variables without adjusting the strength of the element to account for damage. Hence, this may cause an unrealistic tensile force-displacement response. Damage evolution model was referred to be based on linear softening by taking into account the fracture energy data. Besides, in-house developed MATLAB code as regards classical lamination theory would be used to check failure behavior of such laminates with respect to maximum stress, maximum strain, Tsai-Hill, and Hashin failure criteria (Duman 2019). Throughout the numerical analysis, all of the specimens were considered as axially symmetric about its central axis. Lay-up configuration of selected epoxy/carbon fiber composite material used in numerical analysis were represented in Figure 2.

(a)

(b)

(c)

Fig. 2. Stacking sequence of the composite laminates, (a) QI [0/+45/-45/90], (b) SYMB [+45/-45/0/90] s , (c) SYM [+45/30/0/90] s .