5. C modeling#
5.1. Characteristics of modeling#
Modeling with elements of BARRE and CABLE.
5.2. Characteristics of the mesh#
The mesh contains:
2 elements of type BARRE.
60 elements of type CABLE. (15 elements per cable).
5.3. Tested sizes and results#
We test the 1st natural vibration frequency of the cables at the last step of the calculation, PAS =20. As the 4 cables have the same voltage, there are therefore multiple vibration modes. Six multiple modes are tested. It is a non-regression test.
Frequency |
Reference type |
Calculated value |
Tolerance |
1 |
“NON_REGRESSION” |
\(88.5312\) |
default |
2 |
“NON_REGRESSION” |
\(88.8603\) |
default |
3 |
“NON_REGRESSION” |
\(88.9670\) |
default |
4 |
“NON_REGRESSION” |
\(89.0146\) |
default |
5 |
“NON_REGRESSION” |
\(89.0146\) |
default |
6 |
“NON_REGRESSION” |
\(89.0146\) |
default |
Note:
For the record the reference frequency is \(88.728308982\mathit{Hz}\) .
5.4. notes#
The first calculation is carried out with the operator STAT_NON_LINE. The operator CALCUL, which has as inputs the fields of displacement, stress, internal variables and an increment of the movements, makes it possible to know the stiffness matrix of the system. The calculation of vibration frequencies and modes is carried out with the operator CALC_MODES using the stiffness matrix that comes from the operator CALCUL.
The operator CALCULfait assumes the linearity of the fields as a function of the increment of the movements. However, in the case of cables, this hypothesis is not entirely « correct ». This displacement increment should make it possible to approximate the tangent matrix, so it must be « small ». It is determined by a sensitivity study. This is the reason why this modeling is a non-regression test.