SEISMIC BASE ISOLATION OF MULTISTOREY BUILDINGS

Abstract

Man has been exposed to the hardships of earthquakes since the ancient times and due to impossibility to avoid them, he has learnt how to protect himself by improving his construction methods. Broadly speaking, there are two design philosophies for protecting buildings against earthquakes. The first, and most popular, consists in making the building sufficiently strong and ductile, so that it can withstand the earthquake loads without essential damage. The second method incorporates passive or active protection with the aid of various systems used to control the transmission of earthquake shocks to the building and thus reducing the seismic stress occurring in the building. In recent years, application of seismic base isolation has been getting widespread acceptance for protection of buildings from earthquake forces. The concept is based on structural control method which involves in reducing the demand. Behaviour and effectiveness of base-isolated buildings involves several aspects such as frequency characteristics of input base motion, type of isolating device, design parameters of the isolation system, building type and soil conditions. Available literature in the area of seismic base isolation of multistorey buildings is reviewed. Various investigators have performed studies covering various aspects of base isolation such as development of isolator for different design requirements, analytical modelling of their load-deflection behaviour, experimental verification of the analytical models, earthquake simulation testing of base-isolated building models by shake tables, analytical modelling of baseisolated buildings, retrofitting schemes for historical buildings using seismic base isolations, performance of base-isolated buildings in real earthquakes etc. Though base isolation has become popular for buildings of post earthquake importance such as hospitals, communication centres, police stations etc., there are still some issues and gap areas in this approach that need further research and development for its wider application. Afew of these areas have been selected for this study. Behaviour of seismic base isolation of moderate height and tall buildings for different types of base motions has been investigated. The study is conducted in two parts viz. experimental and analytical. The experimental study consists in determining the load-deflection behaviour of laminated rubber bearings which are later used to isolate athree storey one sixth scale reinforced concrete building model. This model is tested under quasi-static loading and for free vibration and harmonic base excitation. The results obtained by analytical models have been compared with the experimental results to verify the accuracy of the analytical models. These analytical models of rubber bearings are later used for analytical study of moderate height and tall buildings. The quasi-static shear tests on rubber bearings gives the stiffness and damping characteristics of the bearings. It is observed that the bearings used in the study possess low damping. Horizontal stiffness and damping shows decreasing trend with increase in shear strain though the rate of decrease is low. The stable hysteresis loop for number of cycles verifies the suitability of rubber bearings, under earthquake type loading. Quasi-static tests on base-isolated model show that superstructure of base-isolated building behaves like a rigid block. The horizontal stiffness of the building is more or less equal to the combined stiffness of isolators provided in the isolation system. Comparison of responses of fixedbase and base-isolated model under harmonic base excitation shows that base isolation with rubber bearings is very effective for high frequency base motions which are filtered due to presence of rubber bearings. Response of the model, considering it as afull scale building, is also computed analytically by modelling the load-deflection behaviour of rubber bearings as (i) combination of linear elastic and viscous damping element and (ii) hysteretic element respectively. The parameters of these elements are taken from the results of quasi-static tests of rubber bearings. Experimental and analytical responses are compared and found to be in close agreement. Analysis of base-isolated buildings is a complex problem due to non-linear behaviour of isolation devices. Several structural analysis and finite element analysis softwares, then available, have been used by different authors in the past. These softwares, in some conditions do not give sufficiently precise results or even fail to analyse the base-isolated buildings. Available software 3D-BASISTABS is found suitable to conduct the analytical study performed in this thesis. The programme assumes superstructure to be elastic and non linearity to be concentrated in the isolation system only. Elements are available in the programme for modelling elastomeric and sliding type bearings. Two reinforced concrete buildings of four and six storeys respectively are analyzed for the study of moderate height buildings. It is well known that in the region of rock or stiff soil, fundamental frequency of conventional moderate height buildings generally fall in the range where the earthquake energy is maximum. Due to quasi-resonance condition, the building attracts substantial earthquake energy. Therefore base isolation is very promising and effective alternative technique for moderated height buildings. Effectiveness of base isolation for these buildings is studied for earthquakes of different characteristics. Effect of bearing parameters such as damping, post-topre yield stiffness ratio and yield-force is studied for these buildings. Amount of damping in isolation system plays a very important role in controlling the seismic response of base-isolated buildings, particularly the base displacement. The seismic response decreases with increase in isolation damping but beyond a certain value called as optimum damping, higher mode participation starts affecting significantly the total response of the base-isolated buildings. Value of optimum damping in isolation system is observed to be mainly depending on the frequency characteristics of the input earthquake motion and is low for earthquakes having high dominant frequencies. Several types of isolators have been developed till now but only few viz. low damping rubber bearings, filled rubber bearings, Teflon-steel sliding bearings VI1 and Friction pendulum system (FPS) have been developed to the stage of practical application. Suitability of these isolators for different conditions is not very well defined and is studied for different type of earthquake motions. Though these isolators are in general observed to be effective in reducing seismic response of moderate height buildings but their effectiveness depends on different design requirements. Rubber bearings are most suitable for high frequency motions while friction type isolators may also be suitable for sites where base motion can have low frequency contents. The buildings with sophisticated electronic equipment and loose contents should preferably be isolated with rubber bearings to avoid transmission of high frequencies to the superstructure. In case of low gap between the adjacent buildings or where, for some reasons, it is not possible to provide large seismic gaps, the FPS type isolators are more suitable. Base isolation method of earthquake resistant design is based on the philosophy that the fundamental period of building is lengthened beyond the range of dominant frequencies of input base motion. The period of isolation commonly adopted is 2.0sec. Tall buildings would already have higher periods and therefore it was earlier thought that base isolation may not be feasible for tall buildings. Three buildings of 10, 14 and 20 storeys, referred as tall buildings in this study, are analyzed to investigate the effectiveness of base isolation for different types of earthquake motions. Base isolation system is considered as combination of low damping rubber bearings and external dampers. It is observed that base isolation results in slight decrease of seismic response but is not feasible for the unstiffened tall buildings. There seems to be a possibility of increasing the effectiveness of base isolation for tall buildings by stiffening their superstructure. The stiffening may result in low fixed-base period and such buildings, if base-isolated may develop smaller seismic response. Another method that can increase the effectiveness of base isolation for tall buildings may be by increasing damping in the superstructure which can dampen the higher mode response. Further increase in the flexibility of base isolation system Vlll may also enhance the effectiveness of base isolation for such buildings. These three strategies are explored in this study. It is observed that base isolation is effective for stiffened tall buildings. Effectiveness is more for high frequency base motions. Stiffening of superstructure of base-isolated building is beneficial in reducing the floor accelerations and interstorey drifts. In case of base-isolated tall buildings, the increase of damping in superstructure from 2% to 20% reduces its seismic response except the base displacement which remains more or less constant. The reduction is more for base-isolated buildings with unstiffened superstructure as compared to that with stiffened superstructure. The effect of increase of damping in reducing the seismic response is not very significant due to lesser participation of higher modes. Increase in the time period of isolation system from 2.0sec to 3.0sec increases the effectiveness of base isolation for tall buildings. Increased flexibility of isolation system generally makes base isolation feasible for unstiffened tall buildings while further increase the effectiveness of buildings with stiffened superstructure. Effect of base isolation for the buildings with fixed-base period close to isolation period is also studied. Base isolation reduces the response of such buildings also but the reduction is not significant. The base isolation of buildings with fixed-base period close to 2.0sec can be made feasible by using isolation system with period of 3.0sec.