Experimental and Analytical Investigation of Progressive Collapse Through Demolition Scenarios and Computer Modeling

Abstract

Within the past 40 years, abnormal loadings resulting from natural hazards, design flaws, construction errors, and man-made threats have induced progressive collapse in structures all over the world. As progressive collapse behavior has become more prominent, it has made the necessity for design and analysis tools evident. In effort to provide one of these tools, Applied Science International, Inc. introduced its Extreme Loading® for Structures (ELS®) software, capable of progressive collapse simulation.

This research evaluates the effectiveness of Extreme Loading® for Structures as an emerging, nonlinear dynamic analysis software package in modeling progressive collapse scenarios. The ELS® software utilizes the Applied Element Method (AEM) of numerical analysis, separating it from other available software packages. The software and analysis methodology's accuracy are investigated through simulation of two structural implosions. Comparing the predicted response to the documented response, each scenario is evaluated by analyzing the material models, failure criteria, local structural behavior, and global collapse behavior.

The two case studies, Crabtree Sheraton Hotel in Raleigh, North Carolina and Stubbs Tower in Savannah, Georgia, each include an experimental and analytical investigation. The experimental investigations include gathering existing structural information, coordinating with the demolition contractor to simulate the implosion sequence, as well as observing and obtaining documentation from the actual event. The analytical investigation utilizes the Extreme Loading® for Structures software to construct a model for each structure, simulate the implosion sequence, and analyze the predicted behavior. To understand the effects of individual modeling parameters on the model's response, a parametric study was completed. Creation of an evaluation matrix allowed for systematic assessment of the parametric study, as well as the individual model's behavior. For the case studies, a completed evaluation matrix for each iteration can be found in the appendix, providing a rough quantification of the accuracy.

Observations from this research show that the software is capable of successfully modeling progressive collapse scenarios. The software allowed for realistic construction of the models and was effective on various levels in predicting the local and global collapse behaviors. Inaccuracies were discovered in each model and were investigated through subsequent iterations of the analysis. A solution was found for some of the inaccurate aspects, while recommendations for future research are proposed to address the others.

Allowing for the quick and effective assessment of structures, the Extreme Loading® for Structures software has the potential to become a valuable tool in design and analysis of structures for progressive collapse mitigation. Through continuous validation and verification, modeling techniques and parameters can be established, providing engineers with confidence when venturing into this relatively new realm. Eventually, the advancement in knowledge and computing integrated into this software could provide invaluable benefit to society, in the form of economic cost and life-safety.