A study of the implications of soil-structure interaction effects on the seismic response of high-rise reinforced concrete buildings

Abstract

<p>In spite of significant improvements in the quantification of site effects in the NBCC (National Building Code of Canada) since the introduction of the foundation factor (F) in 1965, recent parametric studies have shown the 1990 NBCC provisions for site effects to be inadequate. In addition, soil-structure interaction effects are neglected in the current NBCC provisions. The current study, aimed at investigating the implications of these soil-structure interaction effects on the seismic response of high-rise reinforced concrete buildings, is carried out for three cities in Canada, namely Ottawa, Vancouver and Prince Rupert. Soil models are developed to correspond to the soil classifications used to define F in the 1990 NBCC. For each of the three cities, structural models are developed to represent both 20-storey reinforced concrete ductile moment-resisting frames and ductile flexural walls. Three sets of ground motion records are developed to represent the postulated bedrock motions at each of the three cities, based on the magnitude and source-distance combinations dominating the seismic hazard at the respective sites. The computer program FLUSH is used to perform the analyses of the various soil-structure systems. Results from the current study indicate that the code F values generally underestimate the site effects associated with the respective soil deposits, but appear to be reasonably adequate, in most cases, when soil-structure interaction effects are taken into consideration. In spite of some apparent deficiencies in the code F values, the 1990 NBCC design base shear is shown to be conservative for regular high-rise reinforced concrete buildings. Conventional uncoupled analyses are shown to provide estimates of the coupled base shear demand that are too conservative. A simple measure to account for the inertial interaction effects in uncoupled analyses is proven to provide a significant improvement in the prediction of the coupled base shear demand. A simplified approach to estimate the coupled system period is shown to provide a satisfactory estimate of values based on the rigorous, but time-consuming coupled analyses.</p>