Reliability analysis of H-section steel columns under blast loading

Abstract

In order to prevent structural collapse by terrorist attacks, the column behavior is highlighted since when a column is damaged consequently losing its load bearing capacity, failure will spread to the rest of the structure resulting in the collapse of the entire structure. In this paper, using the reliability theory and taking into account the uncertainties associated with blast loading and material properties, a methodology is improved for determining the damage probability of a steel column under various blast scenarios. Damage is determined according to damage index based on axial load-bearing capacity. The Monte Carlo simulation method is utilized using an explicit finite element method to obtain the failure probability. The results are described in terms of damage probability, effects of boundary conditions, effect of type of probability density function and sensitivity analyses. The results show the importance of considering the uncertainties in assessing the damage index of the columns against explosion. Also, it is found that in the flexural mode the probability of damage decreases as the support condition changes from pinned to fixed ends. Furthermore, the safe protective distances are estimated based on the concept of damage probability. Likewise, results for log-normal random variables are relatively identical with truncated normal random variables. The results of the sensitivity analyses show that damage index has the highest sensitivity to peak reflected pressure.