PSI - Issue 27

Aditya Rio Prabowo et al. / Procedia Structural Integrity 27 (2020) 171–178 Prabowo et al. / Structural Integrity Procedia 00 (2019) 000 – 000

174

4

components such as shell plating and stiffeners, etc. have been conducted and uniform strain, the necking, as well as the necking length, have been determined. The definition of failure strain that is recommended from the evaluation of the thickness measurements is presented in Eq. 3 (Ozguc et al., 2005). ε f ( l e ) = ε g + ε e ( t 2 / l e ) (3) where ε e is the strain on necking, ε g is the uniform strain, t 2 is the plate thickness, and l e is the individual element length. The values in the expression are generally obtained from the tensile test. A sample has been presented by Lehmann and Peschman (2002) as a reference to fracture of steel plate in ship collision. Both of the models are built by shell element under integrated Belytschko-Tsay element formulation. The comparison between the damage pattern from FEM analysis and the real accident is presented in Fig. 1. It can be concluded from these figures that the results between FEM and the damage pattern had a good correlation. The damage size also indicated FEM successfully produced a positive result with the damage dimension 7 meters in length, 4.5 meters in width, and 2 meters in depth. Collision energy from FE analysis was compared with the calculation of the empirical method. The results by the empirical approach were obtained from the sum of energy dissipation between energy to destroy the structure and form folding damage (shown in Eq. 1) and tearing damage (shown in Eq. 2).

(a) (c) Fig. 1. Damage patterns of ship collision: (a) and (b) Based on FE analysis; and (c) Obtained from actual collision accident. (b)

Table 1. The verification of collision energy.

Dissipated Energy (MJ)

Vessel Item

Case type

Location

Gap error (%)

Finite element analysis

Empirical approach by Zhang formulae

Ro-Ro vessel – double side structures (DSS)

Sunda Strait, Indonesia

Side collision

6.91344

6.96731

0.77318

The verification result in terms of collision energy between the FEM analysis result and Zhang’s formula calculation also showed good agreement as presented in Table 1. The difference between FEM and the empirical method is below 5%. The comparison results in terms of damage pattern and verification of energy dissipation indicated that the numerical model in the finite element method produced positive results. The configuration and setting of finite element simulation will be implemented on other collision simulation cases. 5. Modeling of extended collision analysis This section presents the numerical simulation of various collision cases. The collision cases are composed based on different regional locations of target points in the first stage and material class in the later stage. The simulation is conducted to observe and evaluate the characteristic of collision energy on the side structure. The dynamic analysis simulations are performed by using a nonlinear explicit finite element code, ANSYS LS-DYNA. The struck vessel is double skin structure (DSS) type Ro-Ro passenger ship Marisa Nusantara. It will be assumed to be entirely standstill. The striking vessel is cargo reefer Qi Hang which will be modeled as a rigid body. The configurations of the deformable Ro-Ro passenger ship are listed in Table 2.