15. The principle dimensions are shown in Table 4. Numerical simulations are conducted
on the three models. The 3-D FE model is made of beam, shell, and point mass elements. It has 14,000 nodes and 40,000 elements. In order to model full-loading conditions, the container mass is modeled by point mass elements and distributed on bulkheads and hulls. In beam modeling, a thin-walled open cross-section and bulkheads necessitate the use of 2-D analysis of the cross-section. The sectional property distribution of the 3-D FE model is calculated by WISH-BSD and plotted in Fig. 16. The accommodation deck and bulkheads induce drastic changes in the sectional properties. Sectional properties are reflected in beam modeling as the solid lines in Fig. 16. The effect of bulkheads is considered by increasing the torsional modulus according to the method by Senjanović et al. (2009b). equation(75) It⁎=(1+al1+4(1+υ)CItl0)ItEq.
Selleckchem MEK inhibitor (75) was derived by Senjanović et al. (2009b). In Eq. (75), the second and third terms are the total bulkhead contribution to hull torsional modulus. The energy coefficients of bulkheads and stools due to warping distortion are calculated using Eqs. (59), (60), (61) and (62) selleck chemical in the paper of Senjanović et al. (2009b). Table 5 and Table 6 show the energy coefficients of bulkhead and stool due to warping. The bulkheads of the shell 3-D model are modified to be stiffer than the original design because the container mass attached to the bulkheads can cause local modes in lower frequency. Consequently, the strain energy becomes larger than that of the original design. Finally, the effect of the bulkheads is considered by increasing the torsional modulus as equation(76) It⁎=(1+0.143+2.160)It=3.303It The effective shear factor is calculated by integrating the shear stress flow.
The shear stress flows evaluated by 2-D analysis are shown as dotted lines in Fig. 17. The distances from the dotted lines to the solid lines show the magnitudes of the shear stresses. Dry mode natural frequencies of the beam models with and without bulkheads and the Cobimetinib mw 3-D FE model are compared. Fig. 18 shows the eigenvectors of the models. The eigenvectors of the beam models are evaluated at the reference axis on the mass center. Table 7 shows the dry mode natural frequencies of the models. Good agreement is obtained in the results of 2-node torsion and 2-node vertical bending. The consideration of the bulkhead plays a role in 2-node torsion. However, the 2-node horizontal bending result shows a difference in the natural frequency and the eigenvectors. Linear simulations are conducted on the three models. Fig. 19 compares RAOs of the models. Heave, roll, and pitch motions at the center of mass are almost the same in all the models, which include only rigid motions. Flexible motions can be compared in modal motions or sectional forces. Small differences between the models are found in flexible motions and sectional forces.