5. C modeling

5.1. Characteristics of modeling

We are testing the DKT model with TRIA3 meshes. The same boundary and symmetry conditions are applied as in models A and B.

The successive states of equilibrium are obtained using an arc-length control method.

In this case ETA_PILOTAGE = \(\frac{p}{{p}_{\mathit{max}}}\)

5.2. Characteristics of the mesh

Number of knots: 121

Number of meshes and type: 200 TRIA3

Since there are fewer points on the TRIA3 mesh type compared to TRIA7 and QUAD9, we refine the mesh (10x10 meshes).

5.3. Tested values

Identification	Instants	Reference	Aster	% difference
Limit point no. 1
DZ	1.03	—0.0131	—0.01314	0.312
Eta_ PILOTAGE	1.03	0.9916	0.9818	0.987
Endpoint No. 2
DZ	1.78	—0.0170	—0.0170	—0.01704	0.208
Eta_ PILOTAGE	1.78	0.15	0.1072	28.55
Endpoint No. 3
DZ	2.34	—0.0140	—0.0140	—0.01435	2.511
Eta_ PILOTAGE	2.34	—0.4000 —0.4916	—0.5461	11.08
Endpoint No. 4
DZ	2.50	—0.0161	—0.0161	—0.01584	1.606
Eta_ PILOTAGE	2.50	—0.6333	—0.6426	1.462

5.4. notes

The calculation strategy used is divided into two steps:

• calculation under imposed load up to \(P=\mathrm{582.N}\) corresponding to 97% of the critical load,

• « imposed displacement » calculation: then, an imposed displacement is imposed using the technique of the arc length imposed on the entire structure (option LONG_ARC in STAT_NON_LINE).

Using the arc-length technique makes it difficult to define the reference value to be entered in the TEST_RESU command, since these values cannot be imposed. To define the reference values, we looked for the values of \(\mathrm{DZ}\) that were as close as possible to those listed in the table in [§2.2] and we reported the values of the control parameter that we had to obtain for the values of \(\mathrm{DZ}\) in question.