7. E modeling#

7.1. Characteristics of modeling#

The modeling is 3D for the concrete ring and CABLE_GAINE for the cable.

Three calculations are carried out, in all cases, the linear friction is taken to be equal to \(\mathrm{0,01}\) and the curved friction to \(\mathrm{0,03}\).

  1. For the first calculation (EVOLNOLI), the cable is of type PASSIF/ACTIF: PASSIF to the group of nodes” ANCR1 “and ACTIF to the group of nodes” ANCR2 “. Charging is carried out in two stages. Initially, the cable is prestressed with CALC_PRECONT and a tension of \({10}^{6}N\) without anchoring recoil, in a second stage, a displacement \(\mathit{DY}=-\mathrm{0,1}\) to the group of nodes” ANCR2 “is imposed.

  2. For the second calculation (EVOLNOL2) the cable is of type ACTIF/ACTIF. The cable is prestressed with CALC_PRECONT and a tension of \({10}^{6}N\) without anchor recoil. The parameter ITER_GLOB_MAXI is set to 15 (instead of 20 for the other calculations) in order to force the automatic subdivision of the calculation moments during step 2 of the power-up. This makes it possible to validate the construction of the multiplier function associated with the load at this stage, which depends on the reaction force of the degree of freedom GLIS on the first anchor node obtained at the end of step 1.

  3. For the third calculation (EVOLNOL3) the cable is of type ACTIF/ACTIF. The cable is prestressed with CALC_PRECONT and a tension of \({10}^{6}N\) with an anchor setback of \(5\times {10}^{-4}m\).

7.2. Characteristics of the mesh#

The mesh contains 20 elements of type HEXA8 and 20 elements of type SEG3.

7.3. Tested sizes and results#

The reference values are either analytical and presented in § 2.2.2, or are derived from the comparison with modeling D in BARRE.

Here we specify the coordinates of the concrete node \(\mathit{N57}\): \((\mathrm{-}3.889,3.889,\mathrm{0,5})\).

\(\mathit{M193}\) is the third concrete mesh starting from ANCR2.

Identification

Reference type

Reference value

Tolerance

EVOLNOLI Mesh \(\mathit{M105}\) - point 1 - \(N\)

“ANALYTIQUE”

“”

960448.709086365

1%

EVOLNOLI Mesh \(\mathit{M96}\) - point 1 - \(N\)

“ANALYTIQUE”

“”

857741.905702382

1%

EVOLNOLI Mesh \(\mathit{M105}\) - point 1 - \(\mathit{EPXX}\)

“ANALYTIQUE”

2.07664585748E-3

1%

EVOLNOLI Mesh \(\mathit{M96}\) - point 1 - \(\mathit{EPXX}\)

“ANALYTIQUE”

“”

1.85457709341E-3

1%

EVOLNOLI Node \(\mathit{N57}\) - \(\mathit{DX}\)

“AUTRE_ASTER”

1.75362396368528E-04

1%

EVOLNOLI Node \(\mathit{N57}\) - \(\mathit{DY}\)

“AUTRE_ASTER”

1.73862669805773E-04

1%

EVOLNOLI Mesh \(\mathit{M193}\) - stitch 1 - \(\mathit{SIXX}\)

“AUTRE_ASTER”

-1.4385853954159E+05

1%

EVOLNOLI Mesh \(\mathit{M193}\) - point 1 - \(\mathit{SIYY}\)

“AUTRE_ASTER”

“”

-8.2927316052597E+05

1%

EVOLNOLI Mesh \(\mathit{M193}\) - point 1 - \(\mathit{SIXY}\)

“AUTRE_ASTER”

“”

-3.4252241782541E+05

1%

EVOLNOL2 Mesh \(\mathit{M105}\) - point 1 - \(N\)

“ANALYTIQUE”

“”

960448.709086365

1%

EVOLNOL2 Mesh \(\mathit{M96}\) - point 1 - \(N\)

“ANALYTIQUE”

“”

906761.8988894981

1%

EVOLNOL3 Mesh \(\mathit{M105}\) - point 1 - \(N\)

“ANALYTIQUE”

“”

918367.3641803192

1%

EVOLNOL3 Mesh \(\mathit{M96}\) - point 1 - \(N\)

“ANALYTIQUE”

“”

906761.8988894981

1%

Using two non-regression tests, it is also ensured that, on the second calculation, the component GLIS of the force at the node ANCR1 does not decrease (in absolute value) between the last moment of calculation of the first phase of tension and the first moment of calculation of the second phase of tension (which would produce an anchoring recoil during this stage inducing the arrival of false results).

Node \(\mathit{ANCR}1\) - FORC_NODA \(\mathit{GLIS}\) Instant 0.9

“NON_REGRESSION”

Node \(\mathit{ANCR}1\) - FORC_NODA \(\mathit{GLIS}\) Instant 0.90625

“NON_REGRESSION”