|
|
||
|
|
|
|
This slide shows Figs 51 (top) and 52 (bottom) from the 2007 paper by Bushnell. Shown here are static and dynamic nonlinear equilibrium states from a 60-degree STAGS model of the PANDA2-optimized stiffened cylindrical shell listed as Case 4 in Table 4.
These two figures pertain to Case 4, Table 4: no Koiter, yes change imperfection, ICONSV = 1. (Table 4 is listed 5 slides ago.) The initial Imperfection shape is as shown in Fig. 49 of the 2007 paper by Bushnell, with amplitude, Wimp=-0.0625 in.
Shown here in Fig. 51 (upper figure) is the deformed state in the panel skin (shell unit no. 1) from STAGS at the highest load factor PA reached in the nonlinear static STAGS run, PA=0.98. This figure shows the STAGS prediction of outer fiber effective stress (psi) in the panel SKIN at axial load, Nx= -3000 x 0.98 lb/in.
Shown in Fig. 52: The large increase in effective (vonMises) stress in the panel skin during the dynamic STAGS run in which the load factor is held constant at PA = 1.0 (the design load factor) is caused by local buckling of the type displayed in Figs. 4 and 48 of the 2007 paper by Bushnell and in the previous two slides.
This local buckling at PA = 1.0, occurring for an optimum design configuration derived by PANDA2 with use of the “no Koiter” option, is the reason that in Table 4 (listed 5 slides ago), the recommended option is to use “yes Koiter”. (See Case 5 in Table 4).
The optimum design obtained by PANDA2 with the "no Koiter" option (the option in which the local skin postbuckling "Koiter" branch of PANDA2 is skipped) is unconservative because the maximum stress in the panel skin is underestimated by the Case 4 PANDA2 model.
Page 23 / 30