Probabilistic assessment of axial load-carrying capacity of FRCM-strengthened concrete columns using artificial neural network and Monte Carlo simulation

Fabric-reinforced cementitious matrix (FRCM) is considered a unique technology for strengthening structural elements, in particular, concrete columns. The present study investigated the axial load-carrying capacity of the FRCM-strengthened concrete columns using a probabilistic approach. For this purpose, a total of 10 columns were numerically simulated, and a comprehensive database was compiled with reference to the simulation results and the experimental data from the relevant literature. The database contained eight input variables, including the characteristics of concrete, reinforcement bars, and fibers. Afterward, the resulting dataset was utilized for training an artificial neural network (ANN) model to predict the axial load-carrying capacity of the column under monotonic eccentric loading. Next, by substituting the ANN into a limit-state function and defining the input parameters as random variables, the problem was transformed into a reliability one. The established reliability model was subsequently solved via the Monte Carlo method, and the results were presented as the exceedance probability of the axial load-carrying capacity. Moreover, by adding a loop to this algorithm, the probability was calculated for each desired value of the axial load-carrying capacity and presented as exceedance probability curves. The study results showed that the exceedance probability dropped sharply as the axial load-carrying capacity increased, so that the probability beyond 930 kN was expected to be no more than 4.51%. Consequently, the effect of four different distribution functions on the exceedance probability curve was examined. The results revealed that the failure probability of the exponential distribution was larger than that of the normal one. Finally, the coefficient of variation (CoV) was used to calculate the expected accuracy of the Monte Carlo simulation in the reliability model.