BEHAVIOR OF INTEGRAL ABUTMENT BRIDGES UNDER TEMPERATURE EFFECT AND SEISMIC EXCITATION

Abstract

Integral bridges being safe and aesthetically pleasing is gaining popularity in most of the countries including India, because of its low initial cost, reduced long-term maintenance expenses, faster construction and better seismic performance. The analysis of integral bridge is much more complicated as bridge deck, piers, abutments, embankments and moreover soil-pile interaction must be considered as a single system. Analysis of integral bridges without considering non-linear backfill and soil-pile interaction is impractical, as in most of the long span bridges soil respond beyond the elastic limit. The length of the integral bridges mainly depends on the pile capacity, soil type and abutment movement due to intensity of temperature and seismic load and other factors. Most of the integral bridges are constructed in non-seismic regions, where the research has concentrated on secondary stresses, mainly due to temperature which govern the integral bridge design. The construction of integral bridges is increasing in India and other places, which are having high temperature variation and also high seismic zones. In integral bridges, the redundancy or static indeterminacy allows the formation of local mechanisms at selected locations for largely unknown seismic inputs. This concept in integral bridges is proved to be an excellent option for seismic prone areas. In regions of high seismicity, seismic displacement demand can be significantly more than the thermal movements. Thus, it is very much necessary to study the capacities of these integral bridges in resisting various levels of temperature and seismic loadings. Most of the bridge agencies use steel H-piles for integral bridges, which have greater flexibility in comparison to concrete piles. In India, most of the integral abutment and deck extension bridges are constructed on bored-cast-in-situ concrete piles. These bridges are located in the regions having high temperature variation and high seismic zones, where the length of bridge is restricted by lateral pile capacity due to temperature loading or seismic loading or a combination of both as mentioned in Indian codes. In this study, the behavior of integral abutment bridges built on cast-in-situ piles are studied for temperature effects and seismic excitations to determine their maximum possible length under different environment conditions. To study the behavior of integral abutment bridge, a three dimensional non-linear finite element model has been developed considering material nonlinearity. Material nonlinearity is considered for soil-pile interaction by using Winkler soil model with non-linear soil springs, which were developed by using the guidelines given by API and Reese. The passive earth pressure behind the abutment wall is modeled by using the design curves given in Canadian Foundation Engineering Manual (CFM) for dense sand and Manuals for the Design of Bridge Foundations (NCHRP) for medium and loose sand respectively. Material nonlinearity for structural members is considered only for piers and piles, which were modeled as 2 noded beam elements. The finite element model developed is verified by comparing the results with the published literatures on temperature effects. Three dimensional models of five span reinforced concrete integral abutment bridge of 130 m long and 12 m wide constructed on cast-in-situ piles is used to study the influence of abutment-backfill soil, soil surrounding the pile, predrilled hole, abutment and pier flexibility, pile type & pile longitudinal reinforcement on the length of the bridge. Non linear static analysis is conducted in both temperature rise and temperature fall conditions until the formation of first plastic hinge in the pile to find the maximum yield displacement capacity of 1.0 m and 1.2 m diameter piles. Non-dimensional curves relating the temperature effect with length of integral abutment bridge are established. in Sensitive non linear dynamic analysis has been conducted by using five different response spectrum compatible time histories in both longitudinal and transverse directions to study the displacement demand and the force distribution in the integral abutment bridge. Non-linear dynamic analysis is too sophisticated, time consuming and also highly sensitive. However, non linear static procedure such as capacity spectrum method and displacement coefficient method are found to be of great interest and as a better alternative to achieve the displacement demand and the force distribution under considered earthquake intensity. A simplified method to find the target displacement is proposed. In this method the capacity and design curves are retained without converting into capacity and design spectrums. The technique to find target displacement in the proposed simplified method is on the conceptual basis of capacity spectrum method which is very much similar to that of equal displacement approximation or displacement co-efficient method explained in ATC-40. The proposed method is validated by comparing it with capacity spectrum method and displacement co-efficient method. Target displacement and base shear obtained by non-linear pushover analysis is validated by comparing the results with nonlinear time history analysis. The best suitable pushover pattern is taken to limit the integral abutment bridge length. The target displacement obtained by pushover analysis for seismic loading is combined with temperature displacement to find the length of integral abutment bridges built on cast-in-situ concrete piles in high temperature variation and high seismic zone.