Assessment of the effect of non-structural walls on the dynamic properties and engineering demand parameters of a case study building

Abstract

When an earthquake occurs, most international design approaches ignore the contribution of non-structural components (NSCs) to the building dynamic properties. This thesis evaluates the effects of the architectural components, namely curtain walls, and masonry walls, on the mode shapes, fundamental periods, damping ratios, and lateral stiffness of a case study building as well as the engineering demand parameters in terms of inter-story drifts and peak floor accelerations under the effect of selected and scaled earthquake records matching Montreal’s uniform hazard spectrum with a return period of 2500 years. First, ambient vibration measurements (AVMs) were conducted on the bare frame composed of the structural components. Then, another series of AVMs were conducted on the full frame when the NSWs were added to the structure. Modal parameters (natural frequencies, mode shapes and damping ratios) were extracted using the Stochastic Subspace Identification (SSI) implemented in the software ARTeMIS®. The modal parameters of the MDE were obtained for the first mode only at the bare frame stage due to lack of sufficient records, and for the first three modes of the full frame stage. Moreover, the floor acceleration amplifications (FAA) were computed for the roof at bare frame stage, and all floors at full frame stage. Two linear-finite element models of the building at both construction stages were developed in the software ETABS® (CSI, 2017) and calibrated according to the AVMs results. Only the mass of the other NSCs and building content was considered in the analysis. It was found that the building fundamental period decreased by -12.4% when the NSWs were added. The mode shapes remained the same at both stages of construction due to the presence of masonry walls only on the ground floor. Damping has been increased from +1.98% (bare frame) to +3.03% (full frame). Moreover, it was observed that NSWs contributed to the reduction of inter-story drifts at all floors in both directions (from -20.5% to -41.6% in the longitudinal direction, and from -25.8% to -67.6% in the transversal direction). In addition, the FAA of the full frame have been shown to be superior to those of the bare frame in all stages in the transversal direction. In the longitudinal direction, the FAA of the full frame are higher than those of the bare frame in the intermediate floors (2nd, and 3rd), and lower than those of the bare frame in the mezzanine, and the upper floors (4th, 5th, and roof). Indeed, it has been demonstrated that the FAA are not linear, and that they are underestimated by the NBC 2015 which may leads to an underestimation of the lateral seismic force of the NSCs by underestimating the amplification of the floor acceleration. Finally, to evaluate the effect of the NSWs separately on the stiffness of the building, two finite element models were taken: the full frame model, and another model without the NSC (masonry walls or curtain walls), then the effect of this NSW was quantified. It was also observed that masonry walls and curtain walls provided an additional stiffness of +20% and +5% respectively in the transversal direction. While in the longitudinal direction, the masonry walls brought an additional stiffness of +16%, and the curtain walls decreased the stiffness of -4% due to their disconnections from the slabs in the upper floors.

Date	27 Apr 2022
Original language	American English
Awarding Institution	École de technologie supérieure
Supervisor	Rola Assi (Supervisor)