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

1776

3

2.1. Ground motions A set of 7 accelerograms for the collapse limit state was chosen to conduct nonlinear time-history analyses from the FEMA P-695 dataset (FEMA, 2008; Kircher et al., 2010), utilizing the PEER Ground Motion Database (Timothy et al., 2013). Table 1 summarises the seven strong motion records used for the E-W direction in the FFE analyses, including magnitude, peak ground acceleration, and the corresponding ID numbering from FEMA P-695 (FEMA, 2008). The accelerograms were modified to match the target spectrum within the period range of 0.34 s to 2.57 s, encompassing the fundamental period of the structure, which is 1.71 s. Fig. 2 illustrates the collection of acceleration response spectra, both original and scaled, along with the scaled average response spectrum. These accelerograms were employed for the FFE probabilistic analysis, with three scale factors: 1.00, 1.25 and 1.50.

Table 1. Accelerogram set.

ID

Event Name

Station

Component ID

Year 1994 1999 1979 1995 1992 1987 1999

Mw

PGA (g)

1 3 6 8

Northridge, USA Duzce, Turkey

Beverly Hills - Mulhol

2 1 2 2 2 2 1

6.7 7.1 6.5 6.9 7.3 6.5 7.6

0.52 0.82 0.38 0.24 0.24 0.45 0.44

Bolu

Imperial Valley, USA

El Centro Array #11

Kobe, Japan Landers, USA

Shin-Osaka

11 17 19

Yermo Fire Station Poe Road (temp)

Superstition Hills, USA

Chi-Chi, Taiwan

CHY101

a

b

Fig. 2. Acceleration Response Spectra: (a) original, (b) scaled.

2.2. Fire loads Using a discrete sampling uniform distribution, three values of fire load density were selected: 600 MJ/m 2 ; 900 MJ/m 2 and 1200 MJ/m 2 . According to the new generation of prEN 1991-1-2 (CEN, 2023), the 80th percentile value of fire load density for densely loaded office (office including minimum 20 % archive’s surface) , is 905 MJ/m 2 . 2.3. Finite element model To simulate the behaviour of MRF under seismic loading in the context of the capacity design, beams are typically modelled with concentrated plasticity using an elastic beam model connected to nonlinear rotational springs at its ends, that are associated with a constitutive material model able to capture the hysteretic behaviour, such as the modified Ibarra – Medina – Krawinkler (IMK) as explained by Elkady and Lignos (2014). This model simulates the cyclic deterioration in strength and stiffness of the structural steel components. However, the modified Ibarra – Medina – Krawinkler (IMK) material implemented in OpenSees cannot be straightforwardly used in the fire analyses. Another way to model the MRF beams is using a nonlinear beam model formulated by means of fiber-based cross sections with distributed plasticity modelled according to the Giuffré-Menegotto-Pinto material, i.e. Steel02 and