PSI - Issue 64

Patrick Covi et al. / Procedia Structural Integrity 64 (2024) 1774–1781 Patrick Covi and Nicola Tondini./ Structural Integrity Procedia 00 (2019) 000 – 000

1777

4

SteelFFEThermal material (Covi et al., 2023; Covi et al., 2024). SteelFFEThermal could be used both for seismic and fire analysis. However, this approach cannot properly take strength deterioration phenomena into account. To overcome this drawback, a comparison between the concentrated plasticity and distributed plasticity models applied the case study was performed to identify the accelerograms and the scaling factors that could minimise the difference between the two approaches. Two frame models were created to represent the RBS beams, as illustrated in Fig. 3: a) an elastic beam model with attached nonlinear rotational springs with assigned modified IMK material at its ends; b) nonlinear Euler-Bernoulli beam elements based on a corotational formulation and the uniaxial SteelFFEThermal material. a b

Fig. 3. RBS beam models: (a) concentrated hinges made of modified IMK material; (b) distributed plasticity model.

Both model columns are modelled using nonlinear beam elements based on corotational formulation and the uniaxial SteelFFEThermal material. Geometric imperfections were included to allow for flexural buckling. Each column was discretized with eight elements, while beam elements were discretized respectively using twelve elements for external beams and fourteen for internal beams to get adequate precision in calculating displacements, stresses, and strains in sections of each member. Fiber sections were selected to define the cross section of the elements. A leaning column was linked to the frame using horizontal “equalDOF” to simulate: (i) flexural continuity of the columns belonging to the gravity frames, and (ii) the P- Δ effects. The leaning column was loaded on each floor with a vertical load equal to half of the building's seismic gravity load minus the portion directly assigned to the MRF. For brevity, one sample simulation is selected as an example to compare the two frame models. The selected seismic action is shown in Fig. 4a. The earthquake, known as the Northridge earthquake, occurred at 4:30 am, January 17, 1994, with a magnitude of 6.7. Fig. 4b-c illustrate the numerical simulation results of the nonlinear dynamic analysis for the selected acceleration time-history. All the columns remained elastic during the seismic event. Similar checks were made for all the selected accelerograms. Based on the comparison results, although the distributed plasticity model cannot simulate the deterioration phenomena, the behaviour is similar between the two models of the frame. Therefore, the distributed plasticity model was used for the FFE analyses. a

b

c

Fig. 4. (a) Selected accelerogram for the FFE simulation; (b) Deformed shape of the frames at the end of the seismic event; (c) Comparison of the two plasticity models.