PSI - Issue 64

Edward Steeves et al. / Procedia Structural Integrity 64 (2024) 1975–1982 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

1980

6

Firstly, a robustness assessment was performed by analyzing the intact and damaged states of the truss bridge. Fig. 2 showcases the damaged structure in SAP2000 at the threshold of collapse, where two member hinges have formed along the top chord of the truss, circled in red. These elements along the top chord experience compression, and upon member buckling, the system will lose its stability and subsequently collapse.

Fig. 2. Progressive collapse simulation of damaged truss bridge shown at the threshold of collapse.

The structure is statically determinate, implying that upon one member failure (two in this case because the geometry and loading conditions are symmetrical), the system will collapse. For both the intact and damaged systems, the end diagonals and the top chords have the highest utilization ratios, causing the governing failure mode to be member buckling. Therefore, to upgrade this bridge, it is logical to build-up the end diagonals and top chords so that the governing failure mode is gross section yielding of a tension member instead, a failure mode that involves significant strain hardening after hinge formation. This upgrading approach should improve the redundancy, and by extension the robustness, by allowing load redistribution after element failure. The damaged bridge was upgraded by adding 7 x 350 mm continuous cover plates to the top and bottom of the back-to-back channels comprising the end diagonals and top chords; the connections were also upgraded wherever they governed over member capacities. The result of the pushdown analysis for the upgraded truss bridge is presented in Fig. 3 where the truss model is shown at the threshold of collapse. The member hinges along the bottom chord circled in green formed first, then initiated strain hardening, eventually causing the member hinges along the end diagonals to form, circled in red, triggering collapse. Noteworthy is the difference in the deflection magnitude at collapse between the damaged and upgraded structures: 44 and 171 mm respectively. The difference in magnitude is displayed visually by comparing Figs. 2 and 3.

Fig. 3. Progressive collapse simulation of upgraded truss bridge shown at the threshold of collapse.

Table 2 presents the structural robustness and structural redundancy indices for the damaged and upgraded structures. Although the upgraded system does not yield a robustness index nor a redundancy index of one, the robustness is increased by 56%, and the redundancy is increased by 1663% (near zero to 0.139). Albeit an