Comportement des piles de ponts sous l'effet multidirectionnel des mouvements sismiques

A comprehensive study has been conducted to investigate the seismic behaviour of bridge columns under bidirectional earthquake ground motions. The study comprises a series of linear elastic and non-linear dynamic analyses on a set of regular bridge structures representative of common type bridges that are encountered the most in North American highways, as well as bidirectional cyclic testing of four half-scale rectangular reinforced concrete bridge columns.In the second part of the study, a new Percentage Rule for the prediction of interacting seismic responses of bridge columns under multi-directional earthquake components is derived for the two sites. A total of 14 regular bridges with different geometrical properties were examined for the two sites. The new rule was developed by comparing the column longitudinal steel ratio required from response spectrum analysis with combination rules to the ratio determined from the results of multiple linear dynamic time history analysis of the bridge structures under bidirectional ground motions. The steel ratios were computed for the bending moment demand from response spectrum analysis using the combination rule with different values of the weighted percentage alpha (0, 0.3, and 1.0) and the skew angle (10° to 45°). These ratios were compared to those obtained from multiple elastic time history analyses to identify the bridge properties that influence the most the combination rule.In this last part of the study, the need for combination rules to account for seismic demand from the simultaneous action of bidirectional horizontal ground motion earthquake components and the current CSA S6 requirements regarding the lower limit of 0.8% on the longitudinal reinforcement ratio and the transverse reinforcement for confinement for bridge columns were examined through bidirectional cyclic tests on four identical half-scale reinforced concrete rectangular bridge column specimens. The prototype structures are regular common two-span, skewed bridge structures designed according to CSA S6-06, and the study focused on the demand from earthquakes expected in Eastern and Western Canada. The column specimens are 1.2 m x 0.6 m in cross-section and 3.0 m tall, with axial loads varying between 4.5 to 9.3% Agf'c. Two specimens were designed for Montreal, QC, Canada (east site) using 0% and 30% combination rules, resulting in longitudinal steel ratios of 0.41% and 0.57%. Two specimens represented columns part of bridges located in Vancouver, BC, Canada (west site), with longitudinal steel ratios of 0.97% and 1.72% resulting from the application of 0% and 40% combination rules. Site specific cyclic displacement test protocols were developed from time history analysis of the bridge structures that were previously performed in the second part of the study.All columns exhibited satisfactory inelastic cyclic bidirectional response up to high ductility levels. For both sites, the Combination Rules used in design had no significant influence on the inelastic cyclic response of the columns. The columns designed for the Montreal site exhibited satisfactory inelastic cyclic performance even if they had longitudinal reinforcement ratio less than the current CSA S6 limit of 0.8%. Buckling and rupture of corner bars were however observed at high ductility levels in two specimens, showing the need for additional restraint for these bars and more stringent requirements for transverse reinforcement in columns subjected to bi-axial loading. For all specimens, the height of the plastic hinge region was approximately equal to the smaller column dimension, rather than the larger one as currently specified in CSA S6. (Abstract shortened by UMI.)The first part of the study was aimed at investigating the adequacy and effectiveness of the 30%- rule, at predicting the seismic demand on bridge columns when subjected to bi-directional earthquake motions. A series of multiple time-history analyses performed on nine generic bridge models to determine the exact seismic critical bi-axial response of the bridge columns under pairs of orthogonal seismic ground motion time-histories. Response spectrum analyses were also carried out, using the acceleration spectra of the orthogonal ground motion components, to determine the maximum bi-axial response in each direction and the critical bi-axial response was estimated using the 30%-Rule, as specified by the CSA S6 and the AASHTO bridge design code and regulations.