Estudo da confiabilidade estrutural no dimensionamento de vigas de concreto armado submetidas à flexão simples

Abstract

The structural reliability study enables the consideration of uncertainties inherent to the various design parameters, in order to determine the security levels on which a certain structure is submitted to, through its failure probability analysis. With the aim of identifying the influence of the variables that characterize the ultimate limit state of reinforced concrete beams subjected to a bending moment, this work proposes a structural reliability analysis of beams designed within Brazilian standards such as NBR 6118:2014 “Design of structural concrete – Procedure", taking in consideration the transformation methods – FOSM (First Order Second Moment Method), FORM (First Order Reliability Method) and SORM (Second Order Reliability Method) – and simulation methods – Monte Carlo simulation and Importance Sampling Monte Carlo simulation. To make this analysis possible, it had been varied the parameters of effective depth (as a function of the beam height), concrete compressive strength, reinforcement steel area and the proportion between live load and total load, on 14-and-19-centimeter-width beams. In addition, it had been developed a reliability-based design optimization, with the aim of obtaining the reinforcement steel area needed to provide the target reliability index predetermined. The FORM and SORM as well as the simulation methods presented satisfactory results. In general, it was observed the reliability index decline with the live load rate and effective depth increase, whereas the steel reinforcement area increase led to a rise in the reliability index. The results also demonstrated that the parameters with higher influence on failure probability were the characteristic bending moment due to live load, whose variation had caused a great range on reliability index results, and model uncertainties in general. The parameters with least influence on the failure probability were the concrete compressive strength, effective depth and beam width. The results demonstrated, therefore, the parameters on which the uncertainty must be taken in consideration and monitored with higher attention, and the parameters on which the uncertainty could be neglected.