Effets du basculement des fondations superficielles sur le comportement sismique des murs de refend en béton armé

The main objective of this thesis was to study the effect of foundation rocking on the seismic behaviour of reinforced concrete shear walls for eastern Canada.A 10-story reinforced concrete building located on a site class C, in Montreal, QC, was considered. The seismic force resisting system of the building consisted of six simple shear walls and was designed using a combined seismic force modification factor RdRo equal to 5,6. The design complied with capacity design principles and was performed according to the requirements of the National Building Code of Canada (NBCC) 2005 and the Canadian concrete standard CSA A23.3-04. The shallow foundations supporting the shear walls were designed for three levels of lateral loads: elastic seismic lateral loads divided by RdRo = 2,0; forces corresponding to the nominal flexural strength of the walls at their bases; and elastic seismic lateral loads divided by RdRo = 5,6.A numerical model based on the concept of Beam on Nonlinear Winkler Foundation was developed to study the soil – structure interaction (SSI). This model can simulate the plastic behaviour of the soil as well as the geometric non-linearity caused by the lift off of the foundation. The model was then validated using a parametric study. Historical and artificial representative ground motion time histories were selected for the seismic analysis. The historical accelerograms were scaled using different methods to make them compatible with the NBCC 2005 design spectrum.Three series of seismic analyses were performed. Nonlinear incremental static (pushover) analyses were first conducted on one of the shear walls of the building in order to assess its characteristics. 2D nonlinear dynamic time history analyses were then performed on the same shear wall. The dynamic analyses showed that smaller foundations generally lead to more pronounced rocking response, which reduced force demand at the wall bases. More specifically, the suitability of the provisions of NBCC 2005 and CSA A23.3-04 for the design of shallow foundations was examined. First, the wall was found to yield at its base without lift off when supported on a foundation designed for elastic seismic lateral loads divided by RdRo = 2,0. In this case, it is not necessary to consider SSI because the foundation is too massive to rock. When the wall rests on a foundation sized using a capacity design approach, the foundation lifts off before the wall yields. Foundation rocking then results in soil yielding and permanent deformations. However, these settlements were not found to be excessive. In addition, foundation rocking can significantly reduce the force demand at the wall base. Finally, when the wall is built on a foundation designed for elastic seismic lateral loads divided by RdRo = 5,6, most of the earthquake energy is dissipated through foundation rocking, which has the effect of significantly damaging the soil under the foundation and therefore threaten the integrity of the wall and its foundation because of large permanent settlements.In the third series of analyses, four different nonlinear dynamic time history analyses were performed to validate the use of a 2D model of a shear wall forming part of a 3D building: 2D analyses; 2D analyses with amplification of the accelerograms to reflect in-plane torsional response of the building including accidental torsion; 3D analyses using only the major principal component of the ground motion; and 3D analyses using both principal components of the ground motions. The results showed that 2D analyses with amplification of accelerograms for torsional response generally greatly overestimated the displacements and forces acting in the structure compared to those predicted by the 2D analyses, the 3D unidirectional analyses and the 3D bidirectional analyses. The comparison between 3D unidirectional and bidirectional analyses showed that applying the major principal component of a ground motion is sufficient to study the overall behaviour of the building. Finally, it was concluded that the use of 2D analyses to analyze the seismic behaviour of shear walls forming part a 3D building is quite acceptable.