Assessment of Nonlinear Approaches for the Seismic Analysis of a Continuous Integral Bridge Frame Structure

Abstract

Nonlinear static procedure by combining the pushover analysis and a technique of computing the performance/target points is a powerful tool for investigating the inelastic behavior of civil structures in last decades. Its earlier application was mainly focused on the analysis of building structures. It was later adopted in the analysis of bridge structures by some researchers. Various nonlinear static analysis approaches are available in the United States and Europe regarding the methods for the determination of the performance points in the capacity curves, obtained from the pushover analysis. However, the investigation of their accuracy and performance is still limited. Therefore, this study focuses on the investigation of different nonlinear analysis approaches, namely, N2 Method in Eurocode 8, the Coefficient Method of FEMA-356, Displacement Modification Method of FEMA-440, Capacity Spectrum Method of ATC-40 and Equivalent Linearization Method of FEMA-440, for the seismic analysis of a continuous integral bridge frame. The nonlinearity of bridge piers is modeled by applying the concentrated plastic hinge beam elements in SAP2000, while the bridge girder is modeled by applying the elastic beam elements. Uniform and modal load patterns are employed. The pushover analysis in both longitudinal and transverse directions is performed. The performance points are obtained and compared to those of the nonlinear response history analysis method. Two intensity levels of earthquakes with PGAs of 0.175 g and 0.35 g, which represent the damage limitation and collapse prevention levels, respectively, are taken as input. The P-A and soil-structure interaction effects are also included in the analysis. Our results show that the differences between the computed target displacements by using all nonlinear static analysis methods and the nonlinear dynamic analysis method in both directions are about 0-26.1% and 1.2-10.8 % for the first and second earthquake levels, respectively. While the differences of the targeted base shears are about 1.5-32 % and 7.1- 24.5% for both seismic levels, respectively. Finally, it is found that the Coefficient and the Displacement Modification Methods lead to the nearest target displacements comparing to those of the reference method, while the other analysis methods can also provide reliable target displacements and base shears for the damage limitation level. However, all nonlinear static approaches underestimate the target base shear in the longitudinal direction for the collapse prevention level.