In previous post, I post S-beam crash from the RADIOSS guide
but I think lack in this case study, so I'll execute this S-beam crash case.
In guide, write this simulation iteration information
First, hourglass is described in the radioss guide as follows.
Hourglass modes are element distortions that have zero strain energy. The 4 node shell element has 12 translational modes, in this modes has 3 hourglass modes (7, 8, 12)
It refers to the unrealistic deformation of elements in response to external forces or internal stresses.
And this phenomenon happens primarily due to inadequacies in the numerical model used torepresent the element's physical behavior.
In the figure above, the solid rectangle at 7 is subjected to a bending moment and deforms into a dotted line shape. However, the energy at the Gaussian point of the element is constant. It's calculated as if something has changed physically but not numerically.
We can solve these problems by using elements with more degrees of freedom or by reducing the size of mesh.
Change the 4-node shell element formulation via ${I_{shell}}$ in HyperCrash's Properties.
In this S-beam Crash example, compare the change in results due to these element formulations, the number of intergraion points through the thickness value(N) of 0 or 5, and the velocity of 5m/s or 10m/s.
The element foumulation can be selected as shown in the picture below. In the previous post, I used 0.
Each formulation is characterized by the following
BATOZ : fully-integrated shell element, no hourglassing
QEPH : reduced integration element with physical hourglass stabilization
Q4 Belytschko : reduced integration element with elasto-plastic hourglass with orthogonality
The time cost was BATOZ>QEPH>Q4, with Q4 being the fastest, but there was an hourglassing effect.
It will be important to choose the right foumulation when analyzing the shell element, and I will consider it further if necessary.
'Structural Analysis > OptiStruct' 카테고리의 다른 글
14_Pendulums(2) (0) | 2023.12.14 |
---|---|
13_Pendulums (0) | 2023.12.13 |
11_Impact analysis (0) | 2023.12.11 |
10_Impact analysis (0) | 2023.12.10 |
9_Contact analysis (0) | 2023.12.09 |