| description abstract | A reliability model was developed and used for the probabilistic assessment of typical steel girder bridge structural systems exposed to blast loads. Blast load scenarios and failure criteria were based on field observations of actual blast-damaged bridges in Iraq, where failure was practically defined as the inability of the structure to allow passage of at least one lane of traffic for emergency use after the blast event. Structural performance was assessed with a large-strain, large-displacement numerical model that allows element disintegration, separation, and contact. As it was found that system failure was governed by girder end failure, a response surface was developed to model critical girder end displacement for use in the limit-state function. Random variables were used to characterize uncertainties in the most significant factors affecting reliability, including blast load, geometric, and material strength parameters. It was found that blast reliability was most sensitive to the number of girders used for design, while other influential parameters were web stiffener width and girder yield strength. Parameters found to be moderately significant were girder web depth and deck thickness, while other design parameters, including girder flange thickness, web thickness, deck concrete strength, and deck reinforcement ratio, minimally affected blast reliability if altered within a reasonable range. Effective increases in steel girder bridge blast reliability could be achieved by using smaller girder spacing, higher yield strength girders with lower depth, and a deck with lower compressive strength but greater thickness. | |
