Nelson Vila Pouca

SIMULATION OF THE CYCLIC BEHAVIOUR OF R/C RECTANGULAR HOLLOW SECTION BRIDGE PIERS VIA A DETAILED NUMERICAL MODEL

In the search for robust constitutive models suitable for reproducing the performance of bridge piers during a seismic event, this paper details the simulation of the cyclic responses of four rectangular hollow section R/C bridge piers. These four R/C bridge piers were built at scale 1/2.5 and tested experimentally. Both tall and short piers are considered, covering situations where bending or shear are of relevance. Furthermore, the four piers were reinforced according to rather different design strategies: (I) the first is a 30-year-old bridge designed without allowance to the seismic action, and (ii) the second is a bridge fulfilling the EC8 provisions. The detailed constitutive model that provides the numerical predictions includes two submodels: one with two scalar damage variables, reproducing the tensile and compressive degradations of concrete, and the other is based on the Giuffre-Menegotto-Pinto formulation, simulating the cyclic behaviour of the re-inforcement. The Damage Mechanics submodel is implemented at the Gauss points of the finite elements that discretize the concrete, whereas the steel submodel is implemented on the 2-noded truss elements adopted for the rebars. A comparison between the numerical and the experimental results is discussed in detail in this paper.

SEISMIC BEHAVIOUR OF A R/C WALL: NUMERICAL SIMULATION AND EXPERIMENTAL VALIDATION

This paper describes the numerical simulation of the seismic behaviour of a mock-up of a six-floor building, constituted by two parallel R/C walls and experimentally tested on a shaking table. Within the scope of an international benchmark the mock-up was submitted to three earthquakes with intensities up to 0.71 g, which induced nonlinear behaviour in the concrete and reinforcement. For the numerical simulations concrete is discretised with 2D finite elements, and its behaviour reproduced via a constitutive model with two scalar damage variables. Steel rebars are discretised with 2-noded truss elements, and their constitutive behaviour under cyclic conditions reproduced by the Menegottb-Pinto model. Specific attention is devoted to Rayleigh damping, focusing on two different strategies: (i) disregarding the damping contribution, or (ii) adopting a damping matrix that takes into account the stiffness changes during the nonlinear analyses. Main results and strategies for simulating the benchmark axe presented, with emphasis on the comparison between the numerical and the experimental results, which show good agreement when the damping contribution is neglected.

EXPERIMENTAL AND NUMERICAL ANALYSIS ON THE CYCLIC BEHAVIOR OF BRIDGE PIERS WITH AND WITHOUT CFRP RETROFIT

The main objective of this paper is to evaluate the cyclic behavior of RC hollow piers, with and without with CFRP retrofit, by comparison of experimental tests results with structural numerical modeling. The retrofit techniques aim to increase the shear strength and the ductility capacity through the establishment of principles and strategies applied in an experimental cyclic campaign of RC hollow piers, carried out in the Laboratory for Earthquake and Structural Engineering (LESE) of the Faculty of Engineering of University of Porto (FEUP). The evaluation and calibration of the efficiency of several retrofit solutions is also performed. The numerical simulations are carried out using two different methodologies: (i) fiber model and (ii) damage model. The fiber models based in a finite element discretiztion with non-linear behavior distributed along the elements length and cross-sectional area, while the damage model is supported on refined finite element (FE) meshes, with high complexity and detail levels in the constitutive laws defined for both concrete and steel. The concrete is simulated with a continuum damage model where several applications for bridges with hollow section piers can be found at Faria et al. [1]. Results of the experimental campaign allow to discuss and conclude about the efficiency of each numerical method, namely regarding the shear strength and the ductility capacity assessment.

NUMERICAL SEISMIC SAFETY ASSESSMENT OF RC BRIDGES WITH HOLLOW PIERS

The seismic damages of reinforced concrete bridges in recent events show that many of them have an inadequate behavior and safety. Therefore, is important to accurately define bridge safety assessment and hence evaluate the accuracy of the analytical methodologies for seismic response of bridges. The main objective of this paper is to present several options for structural simulation with different complexities in order to assess the seismic response of bridges and then use the results for the safety assessment with probabilistic approaches. The numerical simulations are carried out using three different methodologies: (i) plastic hinge model, (ii) fiber model and (iii) damage model. Seismic response of bridges is based on a simplified plane model, with easy practical application and involving reduced calculation efforts while maintaining adequate accuracy. The evaluation of seismic vulnerability is carried out through the failure probability quantification involving a non-linear transformation of the seismic action in its structural effects. The applicability of the proposed methodologies is then illustrated in the seismic analysis of two reinforced concrete bridges, involving a series of experimental tests and numerical analysis, providing an excellent set of results for comparison and global calibration.

Case Studies and Comparative Evaluation of Methods

The methods presented in chapter 3 are applied here to specific case-studies, involving bridges with different length and configuration. The chapter starts with a critical discussion of the basic parameters that influence the applicability of pushover methods. Then, a number of case-studies are presented in a rather uniform and detailed way; they were selected among those available with a view to including at least one application of each category of methods described in the previous chapter and (wherever feasible) to applying two or more ‘simplified’ methods to the same bridge structure. In addition to a number of pushover analyses, all case-studies include also response-history analysis of the inelastic response of the bridge, which serves as a reference for evaluating the results of the various approximate (static) procedures. In the case studies, in addition to the four pushover methods described in detail in Chapter 3, some other variants of the key approaches are also used and comparatively evaluated, so that at the end a more global picture of practically all analysis and assessment techniques available for bridges is provided. To allow for an even broader view on the issues involved and put the purely analytical methods into the proper perspective, the final section of chapter 4 presents an experimental evaluation of analytical methods, i.e. results from analytical methods (response-history, as well as pushover) are compared with those from the shaking table testing (using three shaking tables) of a 1:4 scale bridge model.