A Integrated Platform for Validated Prediction of Collapse of Steel Structures

Research Category: Engineering and Technology
Presenter: Vitaliy Saykin
Faculty Advisor: Jerome Hajjar

Recently, the prediction of collapse of structures has gained growing attention. It is of critical necessity to be able to predict and model structural collapse due to column removal, which usually models blast loads. Accurate and validated structural collapse models are virtually nonexistent which significantly limits the structural engineering community to adequately design against possible blast loads and prevent collapse. The current project aims at creating an integrated platform for validated prediction of collapse of steel structures by developing an approach in FEA that accounts for element softening and its subsequent deletion through a material model in FEA. This material model includes the use of Void Growth Model (VGM) to model the initiation of softening and the Hillerborg model for modeling the softening itself and element deletion. æThe parameters of these models were calibrated to a comprehensive set of experimental test results of circumferentially notched tensile (CNT) coupon specimens. The authors then validated these calibrated models through comparison with different experimental test results, ranging from tensile coupons to beam-column connections. The approach has shown to be very accurate, and with the element deletion, the modeling approach can easily account for complete structural component separation which is critical to be able to model collapse of structures.