Dynamic Performance of Flush End-Plate Beam-Column Connections and Design Applications in Progressive Collapse
Chang Liu, T. C. Fung, Kang Hai Tan
Abstract
Chang Liu, T. C. Fung, Kang Hai Tan
Abstract
An experimental program was conducted in this study to investigate the dynamic behavior of flush end plate steel beam-column cruciform connections following instantaneous removal of a central supporting column. The dynamic tests were performed by using a quick-release mechanism to simulate sudden column-removal scenarios. In the test program, the geometries of the flush end-plate connections remained unchanged but subjected to five different magnitudes of uniformly distributed load. Another five quasi-static tests were also carried out for comparison with the dynamic response and investigation of the dynamic increase factors (DIF). Test results showed that the release-time durations of the column support force were approximately 30 ms for all five dynamic tests. For the same initial applied dead loads, the maximum dynamic displacement was significantly increased compared to the corresponding quasi-static displacement. In order to observe the full dynamic performance of the tested connection until final failure, a general-purpose finite-element software was used to conduct the three-dimensional numerical simulations. Comparison results showed that the numerical model could accurately simulate the monotonic quasi-static test results and the dynamic test results under sudden column removal. Subsequently, a series of parametric studies was conducted to investigate the connection behavior considering the load-release time, different loading methods, and the magnitude of uniform distributed load. By comparing the full dynamic response of the flush end-plate connection with its corresponding quasi-static response, both the displacement-based DIF and force-based DIF could be respectively determined. The paper also addresses which DIF is more appropriate for the design of structures to resist progressive collapse.
OpenAlex reports 78 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
An experimental program was conducted in this study to investigate the dynamic behavior of flush end plate steel beam-column cruciform connections following instantaneous removal of a central supporting column. The dynamic tests were performed by using a quick-release mechanism to simulate sudden column-removal scenarios. In the test program, the geometries of the flush end-plate connections remained unchanged but subjected to five different magnitudes of uniformly distributed load. Another five quasi-static tests were also carried out for comparison with the dynamic response and investigation of the dynamic increase factors (DIF). Test results showed that the release-time durations of the column support force were approximately 30 ms for all five dynamic tests. For the same initial applied dead loads, the maximum dynamic displacement was significantly increased compared to the corresponding quasi-static displacement. In order to observe the full dynamic performance of the tested connection until final failure, a general-purpose finite-element software was used to conduct the three-dimensional numerical simulations. Comparison results showed that the numerical model could accurately simulate the monotonic quasi-static test results and the dynamic test results under sudden column removal. Subsequently, a series of parametric studies was conducted to investigate the connection behavior considering the load-release time, different loading methods, and the magnitude of uniform distributed load. By comparing the full dynamic response of the flush end-plate connection with its corresponding quasi-static response, both the displacement-based DIF and force-based DIF could be respectively determined. The paper also addresses which DIF is more appropriate for the design of structures to resist progressive collapse.
Key concepts: Structural engineering, Displacement (psychology), Parametric statistics, Progressive collapse, Dynamic testing, Dynamic loading, Dynamic simulation, Column (typography)