ABSTRACT
By allowing considerable strength increase without changing dimensions and weight of the column, the utilization of fiber reinforcement polymer (FRP) in the confinement of existing reinforced concrete (RC) columns has proved to be a viable alternative for column strengthening. Furthermore, FRP lightweight facilitates the transportation and application of this material to the column; FRP sheets and wraps are flexible and continuous, allowing its use in places with difficult access, columns of large dimensions and curved surfaces, as in the case of circular bridge piers (Ferreira & Diniz 2015).
Since most of the variables involved in the project (mechanical properties of concrete, steel and FRP, geometric characteristics, model error, loads, etc.) are random, probabilistic methods are required to assessing the reliability of RC columns confined by FRP (FRP-RC). In this study the reliability of circular FRP-RC short columns under concentric loading, with respect to ultimate strength is assessed. Monte Carlo simulation is used in the computation of the attendant probabilities of failure of the selected columns.
Most of the proposed models for the estimation of the strength of FRP-RC columns have been validated by experimental results of cylinders of plain concrete. As a consequence, these models do not take into account the influence of longitudinal and transversal steel reinforcement, which commonly occurs in the problem of strengthening of existing RC columns.
It was observed that the model that best describes the compressive strength of CFRP confined RC columns is Lee et al. model, with data points with less dispersion along the 45° line, representing the ideal model. Special attention is given to the statistical description of the random variable “model error”. A statistical analysis is performed in order to define the “model error”, which is required information in the probabilistic description of the column resistance, as required in the reliability analysis of FRP-RC columns. Figure 1 shows the frequency diagram and superimposed Normal and Lognormal distributions associated to Lee et al. model error. Frequency diagram and superimposed Normal and Lognormal distributions associated to Lee <italic>et al</italic>. model error. https://s3-euw1-ap-pe-df-pch-content-public-p.s3.eu-west-1.amazonaws.com/9781315207681/cd556cd4-4dcf-4efe-8e29-56fc67b8bfbd/content/fig46_1.tif"/>
A contribution to the development of semi-probabilistic design recommendations for FRP-RC columns under pure axial compression is presented. Monte Carlo simulation is used in the probabilistic description of the column strength, and in the estimation of probabilities of failure of the designed columns with respect to strength. Thirty-two circular RC columns confined by CFRP and steel were selected for analysis. The columns have been designed according to the ACI 440.2R (2008) recommendations for strengthening of members subjected to axial force.
Safety levels implicit in the design recommendations of FRP-RC columns designed according to ACI 440.2R (2008) were assessed. The resulting reliability indexes are in agreement with the selected target reliability index for design of new columns as proposed by Szerszen and Nowak (2003). However, it should be reminded that the target reliability index for existing structures is still an open issue, and more discussion is needed on this topic.
