Reliability of short circular RC columns confined by FRP

Abstract

The use of fiber reinforced polymers (FRP) to strengthen reinforced concrete (RC) columns through concrete confinement has proven to be a viable alternative, especially in the case of circular bridge piers. FRP is a composite material that has high tensile strength, low specific weight and high durability, important properties for strengthening RC structures. In spite of these advantages, a large research effort is still necessary to select analytical models to estimate the ultimate stress and strain of RC columns confined by FRP (FRP-RC). Different models with different levels of conservatism have been proposed in an attempt to describe the effects of confinement. In this study five models for FRP-RC columns are evaluated in terms of the stress-strain behavior and ultimate conditions. Since most of the quantities involved in the design of the column strengthening (mechanical properties, geometry, model error, loads, etc.) are random variables, evaluation of the column resistance and implicit safety levels resulting from the assumed design procedures should be performed via structural reliability methods. In this research, special attention is given to the random variables "model error" associated to the ultimate stress and strain of the confined concrete. An experimental database including circular RC columns with longitudinal and transverse steel (stirrups or spirals) confined by carbon fiber reinforced polymers (CFRP) was compiled from the literature, which allowed to obtain the statistics of the random variables model error. The reliability levels implicit in the strengthening procedures suggested by ACI 440.2R (2008) were assessed. To this end, the probability of failure (and corresponding reliability indexes) of 144 FRP-RC columns were obtained by Monte Carlo simulation. It was observed that the corresponding reliability levels are in line with the target values suggested in the calibration of the American code ACI 318