PROBABILISTIC SHEAR STRENGTH MODEL OF REINFORCED CONCRETE COLUMNS CONSIDERING MECHANICAL MECHANISM AND UNCERTAINTIES

In order to overcome the disadvantage that traditional deterministic shear strength models cannot rationally consider the influence of uncertainties, a probabilistic shear strength model for reinforced concrete (RC) columns was proposed by considering the mechanical mechanism and the influence of uncertainties. The analytical expression of the probabilistic shear strength model for RC columns was established firstly, based on the truss-arch model by taking into account the contributions of concrete, transverse reinforcement and arch action as well as the influence of uncertainties. Then the posterior distribution of probabilistic model parameters was determined by combining the Bayesian theory and the Markov Chain Monte Carlo (MCMC) method. Meanwhile, the influences of the prior distribution and the number of updates on the stability and convergence of the posterior distribution of probabilistic model parameters were investigated. Finally, the applicability of probabilistic shear strength model was verified by comparing with experimental data. The analysis results show that the posterior distribution of probabilistic model parameters will be updated gradually with the increase of experimental data. Moreover, the probabilistic shear strength model is not only able to reasonably describe the probabilistic characteristics of shear strength, but also can calibrate the accuracy of traditional deterministic shear strength models.