Posted on


Steel Structures Lab, Department of Civil Engineering, University of Patras, Patra, Greece
The decisions for the post-fire reuse of steel structures are based on simplified criteria although it is known that a steel structure after fire is “different” in terms of material properties and geometry. Furthermore, it is unknown its post-fire performance under seismic loading, when cyclic loading, plastic-hinge formation and other similar effects are developed. This paper presents an initial study on the post-fire cyclic performance of unprotected beam–column connections. The steel connections are exposed to several fire scenarios based on standard and natural fire curves, resulting material properties changes and deformations. The deformations are limited to code defined limits for reinstatement. Their post-fire performance under monotonic and cyclic loading is studied, compared with their pre-fire performance. The investigation is based on numerical simulations and useful conclusions are drawn for the design of moment connections for improved post-fire performance.

Keywords: Post-fire, Steel Structures, Moment Connections, Cyclic performance, Reinstatement
A large number of analytical, numerical and experimental investigations have been undertaken on the seismic performance of steel structures during last decades (Karavasilis et al., 2006, Karavasilis et al., 2010, Karantonakis et al., 2019 etc.). The research has driven to improved earthquake resisting steel structures with limited structural and non-structural damage due to seismic loading, which can be used almost direct after earthquake.
Similarly, extensive research has been undertaken on the performance of steel structures under fire attack. This research has been driven to improved design methods for fire resistance of steel structures Wang et al., 2012, Franssen et al., 2009. However, the post-fire reinstatement has not been widely studied, especially when the structure is build in a high seismicity area, resulting exhaustive loading, plastic deformations etc. The safe post-fire reuse is important for the building owner / user (business continuation, repair cost etc.) as well as for the insurance company which will pay for the insurance compensation.
An important part of fire exposed steel structures do not collapse and suffer limited structural damage, and so they can be reused after a fire. Examples are the fires to One Meridian Plaza, Philadelphia, USA (1990) (USFA-TR-049, 1991), Churchill Plaza at Basingstoke (2005) (Wang, 2002), and Broadgate, London (1990) (Wang et al., 2012; P113, 1991) etc. Especially for Broadgate fire, extensive investigation was undertaken after fire. Many structural elements were not fireproofed and the active firefighting systems were not active as the building was under construction. Even though the fire had long duration (4.5 hours) and high temperature development (near 1000 oC) the performance of the steel structures was sufficient. Due to limited structural damage, the structure repaired in a short period of time.
The reinstatement of steel structures, after a fire event, is related to the post-fire deformations and the residual mechanical properties. According to previous studies, although the mechanical properties of steel are reduced at elevated temperatures, usually they are recovered after cooling down to room temperature (Outinen et al., 2004; Lee et al., 2004; Tao et al., 2013; BS5950-8, 2003; Cooke, 1998; Li et al., 2003). The recovering degree is function of several parameters (maximum temperature during heating, rate of cooling, chemical composition of steel, thermal treatment of steel during production) (Maraveas et al., 2017a). Generally, the after cooling down residual mechanical properties of high strength steel and bolts show higher degradation than ordinary steel, due to thermal treatment of steel during production (Maraveas et al., 2017a; Maraveas et al., 2017b).
The decisions for the reinstatement of steel structures are based on simplified criteria (Tide, 1998; Kirby et al., 1986; Smith et al., 1986; Maraveas et al., 2017a) although it is known that the steel structure is “different” after exposed to fire, in terms of material properties and geometry. Literature (Tide, 1998; Kirby et al., 1986; Smith et al., 1986; Maraveas et al., 2017a) and a design code (BS5950-8, 2003) propose the reuse of steel structures after a fire when the deformations are very limited, and specific limits in terms of displacements and rotations are proposed. A rising question is if the reuse of steel structures according to these simplified criteria is safe for accidental load combinations, like earthquake loading combinations. Steel structures designed for high earthquake loads can survive a fire attack without considerable damage. Due to design against extreme seismic loads, the cross sections are thick and so they have high critical temperature and so high fire resistance. Furthermore, their load ratio is low when only gravity loads are applying. So, according to these simplified criteria, these steel structures can be repaired and reused. Their post-fire performance to earthquake loading has not been studied until nowadays.
Their seismic performance is affected by the following parameters:
• The reduction of the mechanical properties of steel after the exposure to elevated temperatures and cooling down and
• The post-fire residual deformations
Critical structural elements for the performance of steel structures under seismic loading are moment beam column connections, which are severally affected by fire due to high post-fire degradation of mechanical properties of bolts (Maraveas et al., 2017a; Maraveas et al., 2017b; Maraveas et al, 2021a; Ketabdari et al., 2019). Previous research of authors (Maraveas et al., 2021b) investigated the seismic post-fire performance of a number of moment beam-column connection. The conclusion was that the connections with critical design component the bolts, according EN1998-1-8, 2005, were severely affected by fire in terms of moment resistance, stiffness and energy dissipation.
This paper investigates proper methods for improved post-fire performance of such moment connections. Especially, the effect of over-design of connections in terms of bolts’ steel grade or bolts’ diameter.
2.1 Simulation methodology and analysis parameters
In order to predict the post-fire cyclic performance of moment beam-column connections and the fire effect to cycle loading performance the bellow steps are followed:

  1. Development of the connection numerical model, including local imperfections (b/100 according Maraveas et al., 2017c) and residual stresses (according to BSK 99 (Abambres et al., 2016).
  2. Monotonic / Cyclic loading of the connection following a specific applied displacement history for comparison reasons.
  3. Perform coupled thermal – structural analysis. Three sides exposure of beam considered. During thermal analysis, a fire curve (standard fire curve or extreme parametric fires according EN1991-1-2, 2015) is applied. The fire curve is cooling down when the post fire displacements and rotations are near the limits per BS 5950-8, 2003. In order to approach the displacement or rotation limit, trial and correct rounds of analysis are repeated.
  4. For the post-fire analysis, the maximum temperature developed in each node is set up as initial temperature. The post-fire residual geometry from the coupled thermal-structural is used.
  5. The post-fire residual mechanical model for structural steel and bolts is implemented according Maraveas et al., 2017a..
  6. The thermal expansion coefficient is given as zero.

Type of service: Editing
Type of assignment: Research Paper
Subject: Engineering
Pages/words: 4/2200
Number of Sources: 0
Academic level: Doctoral
Paper Format: MLA
Line spacing: Single
Language style: US English