Confiabilidade de pilares curtos circulares em concreto armado confinados com PRFC sob flexo-compressão

Abstract

The confinement of circular reinforced concrete (RC) columns with carbon fiber-reinforced polymer (CFRP) has established itself as an effective technique for strengthening existing columns. Its growing adoption is attributed to its high efficiency and ease of application, among other advantages. Strengthening of RC columns falls within the domain of existing structures, which is distinct from the design of new ones. Given its importance, developing specific technical standards for CFRP-confined RC columns (RC-CFRP) is necessary. These standards should follow a framework similar to that used for new structures, employing semi-probabilistic methods. An essential step in this development is the assessment of the reliability levels implicit in the strengthening recommendations. This process introduces several additional issues: the need for a stress-strain model that represents the behavior of FRP-confined reinforced concrete; statistical description of the design variables, including inherent and epistemic uncertainties; selection of the target reliability index to be achieved; and addressing challenges related to relative eccentricity. In this study, the reliability levels of 540 short RC columns with circular cross-sections, subjected to axial load and bending, are assessed. Initially, the RC columns are evaluated according to the ACI 318 (2014) criteria and strengthened following the ACI 440.2R (2017) guidelines. A CFRP confinement model that explicitly accounts for the presence of transverse steel and model errors associated with estimating ultimate strength and strain is used to calculate the capacity of the CFRP-RC columns, along with modeling the inherent uncertainties in material properties and loads. Monte Carlo simulation (MCS) is employed to probabilistically describe the column strength and load statistics, as well as to calculate the probability of failure. The reliability indices (β) obtained through this approach are analyzed in relation to the target indices (βtarget) established in the literature. Additionally, an evaluation of the influence of each relevant design variable on β is conducted. The obtained β results, ranging from 3.32 to 4.31, comply with the recommendations for both existing and new structures, covering the range prescribed by reference standards. These values reflect the adequacy of structural performance within the established criteria, albeit with variations that indicate the influence of different design parameters. The results of this research support the efforts of standardization committees in the development and improvement of guidelines for strengthening RC columns with FRP.