5. C modeling#

5.1. Characteristics of modeling#

Loading is a combination of traction — compression and flexure.

_images/1000020000000189000000EB5E0D602A193B4B87.png

Figure 5.1-a : m**mesh and boundary conditions**

Modeling: DKT

Boundary conditions: Traction — Compression and Flexion coupling:

  • \(\mathrm{DX}=0.0\) and \(\mathrm{DRY}=0.0\) on the \({A}_{1}-{A}_{3}\) edge

  • \(\mathrm{DX}={U}_{0}\times f(t)\) and \(\mathrm{DRY}={R}_{0}\times f(t)\) on the \({A}_{2}-{A}_{4}\) edge,

where \({U}_{0}\mathrm{=}1.5\times {10}^{\mathrm{-}4}m\), \({R}_{0}\mathrm{=}5.\times {10}^{\mathrm{-}3}\mathit{rad}\), and \(f(t)\) is the magnitude of the cyclic loading as a function of the (pseudo-time) parameter \(t\). Two types of loading are considered:

  • The same loading function \(\mathrm{f1}\) for membrane and flexure (synchronous case):

_images/10000000000001F8000001201AA96929C8D1ABF2.png

Figure 5.1-b : f1 loading function

  • The \(\mathrm{f2}\) function of membrane loading twice as fast as that of flexure (in practice, the membrane frequencies of a slab are higher than those of flexure):

    images/10000000000001f8000001205BAC63A07247F1DC.png

Figure 5.1-c : f2 loading function

5.2. Characteristics of the mesh#

Number of knots: 9 Number of stitches: 8 TRIA3; 8 SEG2.

5.3. Quantities tested and results: first loading (same loading function for membrane and flexure)#

We compare the sum of the forces along the axis \(\mathit{Ox}\) in \(\mathit{A1}\mathrm{-}\mathit{A3}\), the displacements along the axis \(\mathit{Oy}\) in \(\mathit{A4}\), the moments along the axis \(\mathit{Oy}\) in \(\mathit{A1}\mathrm{-}\mathit{A3}\) and the rotations along the axis \(\mathit{Ox}\) en \(\mathit{A4}\) obtained by the multilayer modeling with the ENDO_ISOT_BETON law and by the one based on the law BETON_REGLE_PR, in terms of relative differences; the tolerances are taken in absolute values:

Identification

Reference type

Reference value

Tolerance

PHASE ELASTIQUE \(t=\mathrm{0,25}\)

Relative difference in efforts \({N}_{\mathrm{xx}}\)

NON_REGRESSION

1 10-6

Relative difference in displacement \(\mathrm{DY}\)

NON_REGRESSION

1 10-6

Relative moment difference \({M}_{\mathrm{yy}}\)

NON_REGRESSION

1 10-6

Relative difference in rotations \(\mathrm{DRX}\)

NON_REGRESSION

1 10-6

PHASE ENDOMMAGEMENT \(t=\mathrm{1,0}\)

Relative difference in efforts \({N}_{\mathrm{xx}}\)

NON_REGRESSION

1 10-6

Relative difference in displacement \(\mathrm{DY}\)

NON_REGRESSION

1 10-6

Relative moment difference \({M}_{\mathrm{yy}}\)

NON_REGRESSION

1 10-6

Relative difference in rotations \(\mathrm{DRX}\)

NON_REGRESSION

1 10-6

PHASE DECHARGEMENT \(t=\mathrm{1,5}\)

Relative difference in efforts \({N}_{\mathrm{xx}}\)

NON_REGRESSION

1 10-6

Relative difference in displacement \(\mathrm{DY}\)

NON_REGRESSION

1 10-6

Relative moment difference \({M}_{\mathrm{yy}}\)

NON_REGRESSION

1 10-6

Relative difference in rotations \(\mathrm{DRX}\)

NON_REGRESSION

1 10-6

PHASE ELASTIQUE \(t=\mathrm{2,25}\)

Relative moment difference \({M}_{\mathrm{yy}}\)

NON_REGRESSION

1 10-6

Relative difference in rotations \(\mathrm{DRX}\)

NON_REGRESSION

1 10-6

PHASE RECHARGEMENT \(t=\mathrm{3,0}\)

Relative difference in efforts \({N}_{\mathrm{xx}}\)

NON_REGRESSION

1 10-6

Relative difference in displacement \(\mathrm{DY}\)

NON_REGRESSION

1 10-6

Relative moment difference \({M}_{\mathrm{yy}}\)

NON_REGRESSION

1 10-6

Relative difference in rotations \(\mathrm{DRX}\)

NON_REGRESSION

1 10-6

PHASE DECHARGEMENT \(t=\mathrm{3,5}\)

Relative difference in efforts \({N}_{\mathrm{xx}}\)

NON_REGRESSION

1 10-6

Relative difference in displacement \(\mathrm{DY}\)

NON_REGRESSION

1 10-6

Relative moment difference \({M}_{\mathrm{yy}}\)

NON_REGRESSION

1 10-6

Relative difference in rotations \(\mathrm{DRX}\)

NON_REGRESSION

1 10-6

Diagrams compared efforts \({N}_{\mathrm{xx}}\) based on displacement \(\mathrm{DX}\) imposed for load \(\mathrm{f1}\) :: **

_images/10000201000002D700000236F8397DAA0BC1D6B0.png

Comparative graphs moment \({M}_{\mathrm{yy}}\) based on the rotation \(\mathrm{DRY}\) imposed for loading \(\mathrm{f1}\) :: **

_images/10000201000002D200000235ABEA33B747C9FDE8.png

Comparative displacement graphs \(\mathrm{DY}\) (due to the Poisson effect) for loading \(\mathrm{f1}\) :

_images/10000201000002D800000235CEF8BC5072DBA5CA.png

Comparative rotation graphs \(\mathrm{DRX}\) (due to the Poisson effect) for loading \(\mathrm{f1}\) :

_images/10000201000002ED000002355D8F40C167936A3B.png

5.4. Quantities tested and results: second loading (membrane twice as fast as flexure)#

We compare the forces along the axis \(\mathrm{Ox}\) in \(\mathit{A1}\mathrm{-}\mathit{A3}\), the displacements along the axis \(\mathrm{Oy}\) in \(\mathit{A4}\), the moments along the axis \(\mathrm{Oy}\) in \(\mathit{A1}\mathrm{-}\mathit{A3}\) and the rotations along the axis \(\mathrm{Ox}\) en \(\mathit{A4}\) obtained by the multilayer modeling with the ENDO_ISOT_BETON law and by the one based on the law BETON_REGLE_PR, in terms of relative differences; the tolerances are taken in absolute values:

Identification

Reference type

Reference value

Tolerance

PHASE ELASTIQUE \(t=\mathrm{0,2}\)

Relative difference in efforts \({N}_{\mathrm{xx}}\)

NON_REGRESSION

1 10-6

Relative difference in displacement \(\mathrm{DY}\)

NON_REGRESSION

1 10-6

PHASE ELASTIQUE \(t=\mathrm{0,25}\)

Relative moment difference \({M}_{\mathrm{yy}}\)

NON_REGRESSION

1 10-6

Relative difference in rotations \(\mathrm{DRX}\)

NON_REGRESSION

1 10-6

PHASE ENDOMMAGEMENT \(t=\mathrm{0,5}\)

Relative difference in efforts \({N}_{\mathrm{xx}}\)

NON_REGRESSION

1 10-6

Relative difference in displacement \(\mathrm{DY}\)

NON_REGRESSION

1 10-6

PHASE ENDOMMAGEMENT \(t=\mathrm{1,0}\)

Relative moment difference \({M}_{\mathrm{yy}}\)

NON_REGRESSION

1 10-6

Relative difference in rotations \(\mathrm{DRX}\)

NON_REGRESSION

1 10-6

PHASE DECHARGEMENT \(t=\mathrm{1,5}\)

Relative difference in efforts \({N}_{\mathrm{xx}}\)

NON_REGRESSION

1 10-6

Relative difference in displacement \(\mathrm{DY}\)

NON_REGRESSION

1 10-6

Relative moment difference \({M}_{\mathrm{yy}}\)

NON_REGRESSION

1 10-6

Relative difference in rotations \(\mathrm{DRX}\)

NON_REGRESSION

1 10-6

PHASE ELASTIQUE \(t=\mathrm{2,25}\)

Relative moment difference \({M}_{\mathrm{yy}}\)

NON_REGRESSION

1 10-6

Relative difference in rotations \(\mathrm{DRX}\)

NON_REGRESSION

1 10-6

PHASE ELASTIQUE \(t=\mathrm{2,5}\)

Relative difference in efforts \({N}_{\mathit{xx}}\)

NON_REGRESSION

1 10-6

Relative difference in displacement \(\mathit{DY}\)

NON_REGRESSION

1 10-6

PHASE RECHARGEMENT \(t=\mathrm{3,0}\)

Relative moment difference \({M}_{\mathit{yy}}\)

NON_REGRESSION

1 10-6

Relative difference in rotations \(\mathit{DRX}\)

NON_REGRESSION

1 10-6

PHASE DECHARGEMENT \(t=\mathrm{3,5}\)

Relative difference in efforts \({N}_{\mathit{xx}}\)

NON_REGRESSION

1 10-6

Relative difference in displacement \(\mathit{DY}\)

NON_REGRESSION

1 10-6

Relative moment difference \({M}_{\mathit{yy}}\)

NON_REGRESSION

1 10-6

Relative difference in rotations \(\mathit{DRX}\)

NON_REGRESSION

1 10-6

Comparative force graphs \(\mathrm{FX}\) ( efforts \({N}_{\mathrm{xx}}\)) — ( efforts) — as a function of displacement \(\mathrm{DX}\) imposed for load \(\mathrm{f2}\) :

_images/10000201000002D700000236B23E287185A71AD1.png

Comparative graphs moment \({M}_{\mathrm{yy}}\) depending on the rotation \(\mathrm{DRY}\) imposed for loading \(\mathrm{f2}\) :: **

_images/10000201000002D20000023562D7413FCF27C14E.png

Comparative displacement graphs \(\mathrm{DY}\) (due to the Poisson effect) for loading \(\mathrm{f2}\) :

_images/10000201000002D800000235AE987DD76773BB5F.png

Comparative rotation graphs \(\mathrm{DRX}\) (due to the Poisson effect) for loading \(\mathrm{f2}\) :

_images/10000201000002ED00000235C921A78F1376694F.png

5.5. notes#

According to the preceding curves, it can be seen that the multilayer model with law BETON_REGLE_PR represents the overall behavior of reinforced concrete under bending and traction in a satisfactory manner under load. The same damage thresholds are identified. However, in discharge, law BETON_REGLE_PR follows the same curve as the charge, unlike law ENDO_ISOT_BETON.

The Poisson effect is not modelled by law BETON_REGLE_PR, so we obtain zero rotation in the DRX direction and zero DY displacement.

For the second load, a similar behavior is observed for the membrane-flexure coupling between BETON_REGLE_PRet ENDO_ISOT_BETON under load (elastic response, damage and compression). The same damage thresholds are identified.