Abstract

1 Introduction

2 Literature survey

3 Dataset details and investigation protocol

4 Proposed method

5 Experimental results and analysis

6 Conclusion and future work

References

Abstract

     Forest fire poses a serious threat to wildlife, environment, and all mankind. This threat has prompted the development of various intelligent and computer vision based systems to detect forest fire. This article proposes a novel hybrid deep learning model to detect forest fire. This model uses a combination of convolutional neural network (CNN) and recurrent neural network (RNN) for feature extraction and two fully connected layers for final classification. The final feature map obtained from the CNN has been flattened and then fed as an input to the RNN. CNN extracts various low level as well as high level features, whereas RNN extracts various dependent and sequential features. The use of both CNN and RNN for feature extraction is proposed in this article for the first time in the literature of forest fire detection. The performance of the proposed system has been evaluated on two publicly available fire datasets—Mivia lab dataset and Kaggle fire dataset. Experimental results demonstrate that the proposed model is able to achieve very high classification accuracy and outperforms the existing state-of-the-art results in this regard.

Introduction

     Forest fire can potentially result in a large number of environmental disasters, causing vast economical and ecological losses apart from jeopardising human lives. These fires pose a serious threat to people, wildlife, and the environment. To preserve the natural resources and protect the properties and human lives, forest fire detection has become very crucial. It has lead to increasing number of research explorations in this area around the world. Early and accurate detection of forest fires is essential for mitigating the effect of the fire as once a forest fire spreads to a large area it becomes very difficult to control it and might result in a catastrophe. In its early stage any forest fire is relatively small and easy to control. Fire and smoke detector sensors can easily be installed in indoor environments, it is generally not the case for forest areas. Sensors also require the fire to burn for a while before they can be detected. On the contrary, vision based devices can be used to detect fire in real life and they can be deployed in any area by using different means. These systems are also cheap and easy to install.

Conclusion and future work

     This article proposes a combination of CNN and RNN based deep learning method for forest fire detection. The evaluation of the performance of the present system has been done on two different public datasets. The proposed forest fire detection system outperforms the existing studies in this regard. The present work also overcomes various drawbacks of the existing systems. It is evident from the high classification accuracy of the present system that the present system can be employed to detect forest fires in the real world scenarios. The present work shall provide fresh insight to the researchers in carrying out the new researches on fire detection using computer vision based techniques.

     In future, the attempt will be made to carry out the research work in this problem area by employing other sophisticated deep learning techniques. The plan is also there to develop a fire detection system in non-forest areas, especially fires that occur in residential areas and industrial areas. Other possible future directions of this research work include the exploration of the possibility of employing the proposed model for low resolution satellite images covering large geographical areas.

1- Introduction

2- Prototype (code-designed) RC building

3- Benchmark for computational modeling of RC structures under fire with heating and cooling phases

4- Behavior of the prototype building under fire

5- Computational model for a column part of a structure

6- Parametric study of the residual deformations of RC columns

7- Conclusion

References

Abstract

Reinforced concrete (RC) structures often remain stable under fire, but exhibit damage and residual deformations which require repairs. While repair operations and building downtime are expensive, current fire design approaches do not consider post-event resilience. The first step to enable predicting the resilience of RC structures under fire is to develop capabilities to model the damage of these structures after various fire exposures. This paper focuses on the prediction of the residual (post-fire) deformations of RC columns within a code-designed five-story RC frame building. Computational modeling approaches to capture the fire behavior of the columns are investigated. The models range from isolated columns with linear springs at the boundaries to full building model coupling beam and shell elements, with intermediate approaches. The analyses highlight the critical nonlinear role of the thermal expansion-contraction of the surrounding beams and slabs on the column deformations. Large transversal residual deformations develop particularly in perimeter columns, combined with residual shortening. This invalidates models based on isolated column or 2D frame. A parametric study of the residual deformations of RC columns is then conducted, with due consideration of the 3D restraints and interactions, to investigate the effects of different design parameters and fire scenarios on the residual deformations after a fire event. The results of the parametric study indicate that fire load density and opening factor significantly influence the residual deformations of RC columns, compared to the thermal conductivity of concrete and live loads. This research improves the understanding and provides recommendations for numerical modeling of the effect of fire on the residual capacity and deformations in RC structures.

Introduction

Reinforced concrete (RC) buildings generally exhibit a good structural performance under accidental fire events, as seen in, for instance, the 2005 Windsor Tower fire or the 2017 Grenfell Tower fire. In the two latter events, no global structural collapse of the concrete structure occurred, despite fires raging for hours. Yet, while fire does often not result in global collapse of RC structures, the potential loss due to downtime and repairs may be significant. In many instances, the fire accident is not as severe as in the two aforementioned cases, and a rehabilitation is possible [1,2]. The question of post-fire damage and downtime cost has gained increasing attention due to the requirements for resilience of structures under hazards. To develop optimum provisions to design fire resilient structures, engineers need the ability to accurately estimate the potential economic loss for different design alternatives due to fire damage. This in turn requires the ability to predict the behavior of RC structures under fire, including the residual deformations and residual load-bearing capacity of a structure after fire. Structural members are often tested as independent elements under fire, disregarding global behavior. However, the proper inclusion of structural continuity of fire induced effects is crucial for an accurate evaluation of RC building’s response. The heating of structural members leads to thermal expansion, which may cause the surrounding structure to impose high restraint forces. The significant impact of boundary conditions on the fire behavior of RC structures under fire has been observed in many historical accidents and previous research. In the Katrantzos Department Store fire (an eight-story RC building) in Greece in 1980, the restraints from differential thermal expansion in the structure led to the collapse of a major part of the 5th to 8th floors and the failures of various other floors and columns throughout the building [3]. Extensive research works have demonstrated the considerable effect of restraints on the behavior of RC structural members under fire, based on numerical or experimental studies of RC columns, beams, slabs and walls [4–11].

1- Introduction

2- Materials

3- Experimental program

4- Test results and discussions

5- Conclusions

References

Abstract

High-performance fiber-reinforced cementitious composite (HPFRCC) has potential to greatly improve the fire resistance and seismic behavior of concrete structures. This paper reports an experimental investigation on post-fire seismic behavior of two-bay two-story frames with HPFRCC joints. Four reinforced concrete frames were fabricated; three of them were tested in compartment fire for 60 min. The fire was regulated following ISO-834 temperature curve. Two different fire scenarios (one- and two-bay fire) were investigated. Two frames were made of monotonic conventional concrete; the other two frames had HPFRCC joints. Each frame was tested under a constant vertical load and a pseudo-static cyclic horizontal load with increased magnitude until the frame failed. The effects of the HPFRCC and fire scenarios on the failure mechanism, hysteretic loops, envelope curve, stiffness degradation, and energy dissipation of the frames were evaluated. The experimental results revealed that the fire exposure reduced the load capacity and deformability of the frames. In the two-bay fire scenario, the use of HPFRCC joints increased the post-fire load capacity by 11%, ultimate deformation by 6%, initial stiffness by 30%, and energy dissipation by 21%. The cyclic behavior of the frame in one-bay fire was better than that in two-bay fire. The frames with HPFRCC joints demonstrated better cyclic behaviors than the virgin reinforced concrete frame.

Introduction

Fire hazards continue to occur in civil engineering structures such as buildings, tunnels, and bridges, and cause catastrophic consequences. The fire behaviors of beams [1–3], slabs [4–6], and columns [7] and frames [8,9] were investigated through experimentation and/or finite element analysis. Effects of different fire scenarios, fire duration, and structural design variables on the structural degradation were evaluated [1–9]. It is agreed that fire exposure can reduce the load-bearing capacity of civil engineering structures. In a fire hazard, the elevated temperature reduces the mechanical strengths of concrete [10,11] and steel bars [10,12], as well as the bond between concrete and steel bars [13,14]. In addition, the elevated temperature may cause explosive spalling in concrete [15]. Concrete spalling may expose steel bars to fire and thus accelerate the degradation of the structure. In recent years, different families of high-performance fiber-reinforced cementitious composites (HPFRCCs) have been developed to improve the mechanical performance, resilience and durability of concrete structures [16,17]. Ultra-high-performance concrete (UHPC) is a family of HPFRCC with extreme compressive and tensile strengths, durability and flowability due to the refined microstructure and welldesigned chemistry [18–21]. Typically, UHPC has a very low water-tobinder ratio (w/b < 0.25) and uses finely ground silica sand. Most recently, river sand and high-volume supplementary cementitious materials such as fly ash and/or ground granulated blast slag have been used to reduce the materials cost and carbon footprint [16,17]. However, the very dense microstructure makes UHPC susceptible to explosive spalling at elevated temperatures due to continuous buildup of internal vapor pressure [15].

1- Introduction

2- A typical underground station and typical fire scenarios

3- Test setup

4- Experimental results and discussion

5- Conclusions

References

Abstract

Fire disasters occurring in underground structures generally result in severe damage. In order to investigate the fire performance of typical underground structures, a large-scale fire test on a three-span frame structure was carried out. The tested model was loaded both by vertical and horizontal earth pressures. The applied temperature history, representing the time-dependent fire load, was determined by simulations of fire scenarios in the underground structure with the help of the software FDS. The paper contains test results of the temperature of the air in the furnace, the concrete and the steel bars in the structural model, and the state of deformation of the model. Apparent phenomena, such as cracking and spalling, were observed throughout the whole testing process. The results show that a very dangerous situation of the underground structure exposed to fire may occur during the cooling process. Curling of the top slab was observed during the heating process. More attention in fire-resistance design must be paid to the columns. The fire test results are useful for validation of numerical models and for further fire-resistance research of underground structures.

Introduction

Fire disasters in underground structures have generally resulted in severe damage of structures [17,5,10,30]. As indicated by performancebased fire resistance design, research results for ground structures cannot directly be used for underground structures, because of different fire scenarios and of different restraint conditions and loads [4,9,7,2]. During the last decades, a large number of simulations of structures subjected to fire, using the Finite Element Method (FEM), were reported in the literature [13,18]. In particular, fire scenarios, i.e. temperature fields to which the underground structures were assumed to be subjected and their performance under the resulting temperature loads, were thoroughly investigated. Savov et al. [22] have studied the behavior of shallow tunnels, subjected to fire loads, by simulating the structure with the “beam-spring” model. Concrete spalling of the tunnel lining was simulated and its influence on the collapse of the structure was analyzed, considering the failure mechanism of beams, modeled by the FEM. This permitted prediction of the failure mode of the tunnel structure. Ring et al. [19] have simulated the behavior of underground tunnels subjected to various fire scenarios, considering different material parameters and structural models. They have performed research of the spalling problem and on the effect of load-induced thermal strains (LITS). Amouzandeh [1] has carried out a series of FE simulations, including treatment of the burning process of fire in tunnels by means of CFD (Computational Fluid Dynamics) and analyses of the structural performance. Their calculations and analyses revealed that the fire resistance of the tunnel with a circular cross-section is greater than the one of tunnels with a rectangular cross-section. Zeiml et al. [29] have investigated the safety of an underground frame structure subjected to fire, with the help of the FEM.

1- Introduction

2- Experimental programme

3- Observations during fire exposure

4- Discussion of results and analysis

5- Conclusions

References

Abstract

This paper presents experimental results of the structural performance of fire-damaged continuous reinforced concrete (RC) T-beams subsequently strengthened with externally bonded carbon fiber reinforced polymer (EB-CFRP) sheets. A series of seven specimens were tested with different fire exposure time and subsequent strengthening techniques. Experimental results showed that both the unexposed control beams and unstrengthened fire-damaged continuous RC T-beams exhibited flexure failure modes and significant redistribution of moment between hogging and sagging regions. The fire-exposed beams had notably degraded strength and stiffness attributable to the internal temperatures exceeding 500 °C for some portion of the exposure. The EB-CFRP strengthening of fire damaged beams was shown to mostly mitigate the effects of fire exposure. The EB-CFRP retrofit measures successfully restored the virgin capacity of the beam and were sufficient to also restore most of the lost initial stiffness. Simple prediction using plane sections analysis and the assumptions of the 500 °C isotherm method [1] were shown to accurately predict the behaviour of the fire-damaged specimens.

Introduction

Elevated temperatures associated with fire exposure cause severe damage to reinforced concrete (RC) structures, resulting in loss of strength and stiffness and the development of relatively large permanent deformations during and following exposure. The behaviours are attributed to the degradation of mechanical properties of both concrete and steel reinforcement and the redistribution of stresses within the beam after fire exposure [2]. Externally bonded carbon fibre reinforced polymer (EB-CFRP) composites are one proposed technique for repairing fire-damaged RC beams. In addition to their acknowledged advantages of having a high strength to weight ratio and good durability [3], EB-CFRP sheets are also easily applied to structures having varying geometry. There exist a number of guide standards for design and installation of EB-CFRP for RC structures – perhaps the most commonly cited is ACI 440.2R-17 [4] which will be referred to in this paper. Countless studies have investigated the flexural behaviour of RC beams strengthened with EB-CFRP sheets under ambient conditions. Results show that EB-CFRP sheets are able to enhance the flexural capacity of beams but decrease the available ductility compared with unstrengthened control beams. Both effects result from the increased effective longitudinal reinforcing ratio afforded by the EB-CFRP. The majority of study on EB-CFRP as a flexural retrofit measure has been conducted on simple span beams and slabs. In experimental studies of continuous RC beams, improved flexural capacity as a result of the presence of EB-CFRP is reported. However, in continuous beams having discrete EB-CFRP applied in regions of high flexural stress, a primary observed mode of failure is the brittle peeling of the concrete cover at the terminations of the CFRP sheets (so-called “end peel” failure) [5,6]. This failure mode, while simple to mitigate in determinate structures [4], is difficult to address in continuous structures in which there is a stress reversal at the girder faces, the point of inflection is not fixed, and moment redistribution is likely. Aiello et al. [7,8] investigated the ability of RC continuous beams strengthened with CFRP to redistribute moment and found that the percentage of moment redistribution of the analysed beams reaches 20%. ACI 440.2R17, for instance, permits redistribution up to 20% but also cautions that the redistribution that can be relied upon falls with increasing EB-CFRP reinforcement. The adoption of hybrid (carbon/glass) FRP sheets has been shown to improve the ductility that may be achieved in FRP-repaired continuous RC beams [9].

Convolutional neural networks (CNNs) have been recently applied to tackle a variety of computer vision problems. However, because of its high computational cost, careful considerations are required to design cost-effective CNNs. In this paper, we propose a CNN inspired by MobileNet for fire detection in surveillance systems. In the proposed network, color features emphasized by the channel multiplier are extracted through depthwise separable convolution, and squeeze and excitation modules further increase the representation of the channel-wise convolution. Custom Swish is used as an activation function to limit exceedingly high weights from the effects of the channel multiplier. Our proposed network achieves 95.44% accuracy for fire detection, which is higher than those achieved other existing networks. Furthermore, the number of parameters used is 38.50% fewer than that of MobileNetV2, the smallest among other networks. We believe that using the proposed CNN, CNN-based surveillance systems could be implemented in lightweight devices without using expensive dedicated processors.

Introduction

Fires can occur anywhere, at any time, and if they are not detected early, they can cause severe damages to property and people. Surveillance systems consisting of multiple CCTVs can be very useful in detecting fires because they are designed to monitor the surroundings 24 hours a day. Furthermore, they can be very useful to monitor fires in a wide range of areas, including inaccessible areas. Consequently, there has been a huge demand for intelligent video-based fire monitoring systems that can alert people to respond quickly by processing and analyzing video streams in real-time. A video-based fire detection system can inform an operator by analyzing videos from CCTVs without using heat, smoke, or flame sensors. Owing to the significant development of video analysis, video signals from CCTVs can be automatically analyzed and can provide alarms to surveillance personnel to enable quick response. Traditional vision-based fire detection methods use handcrafted features, such as color, motion, and texture. Prior studies [1]–[4] detected fire by making full use of color features because fire is generally brighter and has higher contrast than other objects. Ko et al. [1] detected specific fire regions from their color and, then employed a model using wavelet coefficients to detect fire with a support vector machine (SVM) classifier.

Provisioning services

Regulating services

Cultural services

Conclusions

References

A numerical procedure for the thermal analysis of RC slabs at elevated temperatures is the subject of this study, where the moisture increase due to the so-called moisture clog occurring in the cooler parts of reinforced concrete (RC) slabs is introduced as well. Starting from a systematic investigation on the existing concrete constitutive laws available in the literature and in the codes, improved concrete constitutive laws are proposed to describe concrete thermal and mechanical properties at elevated temperatures. The proposed laws – called also models – are validated against well-documented full- and small-scale tests on simply-supported RC slabs. Parametric analyses on the behaviour of RC slabs in fire are carried out as well, to clarify the role that the different constitutive laws may have in the numerical prediction of RC slabs behaviour under fire conditions.

Introduction

In recent years, several full-scale fire tests of reinforced concrete (RC) slabs were conducted to investigate the influence of the tensile membrane action on the behaviour of floor slabs. However, due to the huge costs and complexities associated with the full-scale fire tests, a number of numerical models have been developed by various researchers to predict the fire response of concrete slabs [1-5]. When modelling the behaviours of the slabs in fire, it is very important to accurately and efficiently determine the temperature distribution within the slabs. So far, two main calculation methods have been commonly used to predict the temperature distributions of RC slabs: a combined heat and mass transfer analysis and a heat transfer analysis only. In 1998, Hurst and Ahmed [6] developed and numerically solved a mathematical model to predict heat and mass transfer in concrete structures subjected to fire. The model simulates the changes in pore pressure, temperature and moisture with corresponding changes in concrete properties and has also been able to predict moisture clog. In 2001, Tenchev et al. [7] proposed a comprehensive coupled heat-and-mass transfer model to predict the pore pressures and the moisture migration within concrete at elevated temperatures. In 2003, Gawin et al. [8] developed one mathematical model to predict the pore pressure, moisture transfer and evaporation within concrete at high temperatures. In 2009, Dwaikat and Kodur [9] developed one-dimensional numerical model to predict the fire-induced spalling and pore pressure in concrete structures based on principles of mechanics and thermodynamics. Although these models are capable of predicting moisture concentrations and pore pressures, they are complex and expensive to process. For instance, moisture evaporation or migration is a complex process and depends on many material properties, such as the porosity and permeability of concrete, which are not well known at elevated temperatures. Thus, very often the use of simpler, faster models, which are capable of predicting temperatures with reasonable accuracy, is valid. This is especially true if concrete spalling is unlikely and the moisture can migrate to an unheated surface and escape. As a matter of fact, under physically realistic conditions, if only the temperatures of slabs are required for the structural analysis, a simple approach, in which the effect of moisture is treated explicitly, can be used to predict the temperature distribution of the slabs in fire. For instance, Wickstrom [10] used a modified enthalpy curve applying an average slope to take into account the effect of moisture subjected to evaporation in concrete. Huang et al. [11] proposed a model based on the theoretical heat and mass transfer in concrete. Lamont et al. [12] developed one adaptive heat transfer program HADAPT and considered moisture evaporation by using an enthalpy method.

Abstract

This paper investigates the effect of fire on the performance of geosynthetic reinforced soil bridge abutments using experimental tests and finite element analyses. Experimental programs were comprised of a series of tensile strength tests at elevated temperatures and fire resistance tests, which were performed on a physical model. Findings revealed the adverse effect of fire on geosynthetic reinforced soil bridge abutments when fire duration exceeded 60 min. Results show that the depth within the backfill affected by the fire is approximately 50 cm.

Introduction

In recent years, the use of the geosynthetic reinforced soil (GRS) technology for bridge abutments has been recommended because it has advantages over conventional methods. The GRS bridge abutment system includes a segmental geosynthetic reinforced soil wall with a bridge seat (sill) placed on the top of it. The stability of these structures depends on the mechanical properties of the reinforcing elements as well as their interactions with the soil. Fig. 1 shows a typical GRS bridge abutment system with modular concrete block facing Geosynthetic reinforcements such as geotextiles and geogrids are made from synthetic polymers and mechanical properties of the polymers change with increased temperatures. Nonlinear increases in creep, a significant reduction in tensile strength, increased failure strain, increased degradation, a reduction in the modulus of elasticity, and a reduction in surface hardness are some of the consequences of increased temperatures on the properties of these types of material [1–8]. Few attempts have been made to study the effect of temperature distribution on reinforced soil structures (due to ambient temperature variations). Segrestin and Jailloux [9] investigated the effect of temperature variation on the geosynthetic aging and discovered that in a reinforced soil structure, the temperature within the backfill varies to a depth of 10 m. A seven-year observation of a reinforced earth structure on the M25 motorway at Waltham Cross, UK, carried out by Murray and Farrar [10]. Their observation showed that 0.3 m behind the facing, soil temperature was relatively close to ambient temperature and after a distance of almost 4 m from the nearest external boundary, the soil temperature was constant. Kasozi et al. [8] studied numerically the temperature distribution in a mechanically stabilized earth wall structure in Las Vegas, NV using field data from the Tanque-Verde MSE wall in Tucson, AZ. Based on their study, the overall average temperature within the backfill was much higher than the highlighted test in ASTM D6637 [11].

Stainless steel has the advantages of good appearance, excellent mechanical properties, strong corrosion resistance, easy maintenance and low life-cycle cost. It is widely applied in the field of construction and has wide application prospects. Stainless steel structures have their irreplaceable applicability in marine engineering. Meanwhile, for the architectural constructions in acid-rain-prone areas, because of its superior corrosion resistance, stainless steel structure has excellent performance in saving building maintenance costs and prolonging construction life-cycle. Stainless steel materials have been gradually developed from decorative components to structural load-bearing main components, which have been widely used in civil engineering [1, 2]. However, in recent years, building fires have occurred frequently, and unprecedented challenges are being faced with regard to the safety of building structures in fire. As a building structural material, stainless steel structures or components often do not use any fire prevention measures due to attempts to obtain a good appearance effect. Therefore, the behaviour response and mechanical behaviour of stainless steel structures in fire are particularly important. Alternately, the structure in fire often has a strong integrity. There are quite complex interactions and internal force redistributions between each component. Therefore, it is of great theoretical value to study the fire resistance performance of stainless steel columns with constraints. 17There are many studies on the behaviour response and fire resistance of constrained carbon steel 18 columns in fire. The main contents are the mechanical properties and buckling temperature of constrained steel columns in fire. Neves et al. [3, 4] used the ZWAN program to carry out numerical simulation analysis of the fire resistance of axially constrained steel columns, and a simple analysis model of the fire resistance of steel columns is proposed. Ali et al.

Over the past several years, research has identified firefighters to be an occupational group at increased risk for suicide (see Stanley, Hom, & Joiner, 2016, for review). One study of 1,027 male and female career and volunteer firefighters throughout the United States (U.S.) found rates of suicide ideation and attempts occurring throughout one‘s career as a firefighter (i.e., career rates) to be 46.8% and 15.5%, respectively (Stanley et al., 2015). Similarly, a separate investigation of female firefighters found elevated rates of suicidality, with 37.7% reporting career suicide ideation and 3.5% reporting a career suicide attempt (Stanley et al., 2017a). Moreover, converging data from the U.S. Centers for Disease Control and Prevention (CDC) suggest that firefighters are at increased risk for suicide (McIntosh et al., 2016). The CDC found that workers in protective service roles, including firefighters, demonstrate elevated rates of death by suicide compared to other occupational groups (McIntosh et al., 2016; see also Tiesman et al., 2015). Thus, together with research demonstrating elevated rates of mental health disorders associated with increased suicide risk among firefighters (Carleton et al., 2017), research into factors that exacerbate and mitigate suicide risk among this population is deserving of increased empirical inquiry. Indeed, the occupational responsibilities of firefighters include routine exposure to events that may pose a substantial risk for serious injury or death, such as running into burning buildings, extracting car accident victims from the side of a busy highway, and recovering dead bodies, some of which are themselves suicide fatalities (see Kimbrel et al., 2016). These traumatic exposures could, in part, contribute to a diagnosis of posttraumatic stress disorder (PTSD; American Psychiatric Association, 2013). Incidentally, to date, four separate investigations have implicated PTSD symptoms in the pathogenesis of suicidal thoughts and behaviors among firefighters (Boffa et al., 2018b, 2017; Martin et al., 2016; Stanley et al., 2017b). However, beyond on-the-job traumatic exposures, there are many other occupational stressors experienced by firefighters (Henderson et al., 2016). These stressors include sleep disturbances, due in part to long and erratic shift schedules (U.S. Department of Labor, 2015), marital and family stress (Sanford et al., 2017), and harassment (Griffith et al., 2016). The mosaic of occupational stressors experienced by firefighters may, in part, contribute to elevated rates of suicidal thoughts and behaviors observed across multiple studies (Stanley et al., 2017a, 2015). Nevertheless, research examining the effects of occupational stress on suicidal and suicide-related symptoms among firefighters is scarce, with a few exceptions.

The last decades, the post-fire performance of structural steel elements has attracted many researchers. Although the fire safety of a structure is of paramount importance, the reinstatement of fire damaged structures is in the center of interest nowadays. Since 1960’s, the research is focused on the mechanical properties of the material as well as the holistic behavior of the structure, taking into consideration both low strength [1-7] and high strength steels [8-11], while cold-formed [12] and stainless steel [13] have gained considerable attention as well. In particular for cold-formed steel, distinguish from hot-rolled sections is deemed mandatory, taking into consideration the manufacturing process and the sensitivity to stability phenomena. Despite the recent interest in the post-fire performance of steel structures, an important study published by Kirby et al. [4] in 1986 contributed to the onset of several specifications. Apart from steel features, elevated temperature properties of wrought iron have been documented therein; furthermore, an extensive investigation of the residual properties of cast iron was conducted recently by Maraveas et al. [14]. The mechanical behavior of connections also plays a substantial role in the heating or cooling phase of an axially restrained member. For this purpose, many experimental studies deal with high strength bolt failure after exposure to high temperatures and subsequent cooling down [15-20], to identify both the range and the causes of deterioration. In general, although the mechanical behavior of both structural steel and bolts depends on various factors (such as the steel type, manufacturing process, heating temperature and duration, cooling rate, etc.), the majority of studies indicate that insignificant strength reduction was observed after heating to temperatures up to 500oC. Also, valuable results were observed via experimental fire development on full-scale steel structures. Thestorey steel framed building, while a cold-formed steel portal frame was tested by Johnston et al. [22]. Both of the studies above, which can be observed in Fig. 1, adequately represent the true behavior under fire conditions, aiming to the rational guidance establishment. Despite the importance of fire design of steel structures, specific guidelines have not been established by the current design codes for the determination of the remaining capacity of steel members after the fire event, with the exception of some recommendations proposed by the British Standards [23], the updating of which is essential since it regards limited steel types and specific fire temperatures. In addition, it is worh to consider that steel members perform differently when as part of a structure, owing to certain force and stress redistribution or materials over-strength, compared to the one of individual members; an important detail which is not taken into account by the present regulations. A characteristic example, which has been verified in the literature through both experimental and analytical studies, is the beneficial effect of secondary structural elements on the overall resistance of the structure, even for cold-formed structural systems.

Nowadays, in the light steel framing construction industry, individual profiles with different cross-section shapes (C, U, R, etc.) are used to fabricate built-up members. A built-up coldformed steel (CFS) structural member can span more distance and present higher load bearing capacity and torsional stiffness. Moreover, usually, built-up members are symmetric, eliminating eccentricities between shear and gravity centers, leading to improved member stability. Built-up cross-sections are built fastening individual profiles with self-drilling screws or alternatively using seam weld (Stone and Laboube, 2005; Li et al., 2014; Wilson and Guzmán, 2011). Research has been focused mainly on open sections such as plain and lipped channels, with simple and complex edge stiffeners, with and without holes and angles (Young and Rasmussen, 1998a; Young and Rasmussen, 1998b; Kesti and Davies, 1999); however, in the past few years, some research on the behaviour of built-up CFS members has been conducted (Georgieva et al., 2012; Yuanqui et al., 2014; Young and Chen, 2008). Currently, the effective width method (EWM) is used worldwide, and the direct strength method (DSM) (Schafer, 2008) is used in North America, for design of CFS structural elements. Design guidelines in EN 1993-1-3 (2004) for built-up members are still vague. For instance, the EN 1993-1-3 (2004) only predicts that the buckling resistance of closed built-up members should be determined using the buckling curve b in association with the basic yield strength fyb and buckling curve c in association with the average yield strength fya if Aeff = Ag. In fire situation, the design methods presented in the EN 1993-1-2 (2005) for hot-rolled steel members are also applicable to CFS members with Class 4 cross-sections, establishing the same reduction factors for the mechanical properties of hot-rolled and CFSs. Some studies show that the reduction factors for CFS (Outinen, 1999; Kankanamge and Mahendran, 2011; Ranawaka and Mahendran, 2009a) are different from those prescribed in EN 1993-1-2 (2005). Also, the EN 1993-1-2 (2005) for Class 4 cross-sections limits the critical temperature to 350°C, which may be overly conservative if the load ratio is not considered (Ranawaka and Mahendran, 2009b; Heva, 2009). Moreover, there are no specific design guidelines regarding the influence of restrained thermal elongation in the overall behavior of CFS columns in fire. Several studies on this matter were already conducted but for heavy hot-rolled steel columns (Franssen, 2000; Ali and O’Connor, 2001). The great majority of the conducted research on the fire behaviour of CFS columns has been focused on the local and distortional buckling phenomena using both experimental and numerical analysis (Feng et al., 2003a, 2003b; Lee, 2004; Ranawaka and Mahendran, 2006).