Introduction
Over the past 25 years, the incidence and consequences of natural disasters seem to have increased, including cataclysmic events such as the 2004 Great Sumatra-Andaman earthquake and tsunami that killed almost a quarter million people in 15 nations. As an example of this incidence, Fig. 1 shows the number of earthquakes per year disaggregated by human development index (HDI); the last few years have been particularly damaging in developed countries (New Zealand and Japan in 2011, in particular). In 2011 alone, the UN Office for Disaster Risk Reduction reported that natural disasters resulted in $366 billion in direct damages and 29,782 fatalities worldwide [
1]; average annual losses in the United States due to natural disasters amounted to about $55 billion. Some argue that these losses are coupled to continuous concentrations of population, energy, economic and political power in locations of high risk of natural disasters [
2]. While there is an array of reasons for these disasters, this escalation is also due to certain hazardous events becoming more intense (as a result of climate change effects) but mostly due to decisions that societies and communities have made (or avoided making) that affect the vulnerability of the built environment and social systems [
3].
It is also clear from recent disasters, such as Superstorm Sandy [
4], that the conventional approach of limiting damage through passive means (i.e., land use and building codes) and some active processes (i.e., disaster planning and disaster exercises) are not particularly effective. To stem losses in the future, it will be necessary to both ramp up conventional approaches and develop new ones, including decentralizing response and maximizing the value of network interdependencies [
2]. A common thread through these approaches is the need to both educate the public about risks from natural hazards and involve the public, as well as private, public and government organizations at the local level in community resiliency efforts. Educating and engaging communities in resiliency efforts is difficult because disasters are complex systems and affect multiple critical infrastructures in addition to the human populations, either directly or through cascading and feedback effects [
5].
This paper explores the role of steel and composite structures in improving community resiliency against natural disasters, beginning with some remarks about resiliency and sustainability of the steel industry as a whole, before moving to some examples of how existing and new steel and composite structures can contribute to a more robust infrastructure. The paper is meant to serve as an introduction to six more technical papers that follow which describe specific approaches to improve the resilience of particular steel and composite structural systems. The success of these approaches should be judged against the criteria outlined in the next section.
Natural disasters, sustainability and resiliency
It is easy and fashionable today to say that a certain type of structure or construction material is sustainable and resilient. In the technical literature these arguments are almost invariably made without providing any context beyond a few simplistic comparisons and generalities. To go beyond that superficial approach, a few definitions need to be explicitly stated here first. The intent is to show that indeed steel structures can be classified as resilient and sustainable with respect to direct natural disasters. Direct disasters are defined as having a sudden onset and immediate consequences (i.e., earthquakes, hurricanes, and volcanic eruptions). This is in contrast to indirect, or stealth, natural disasters [
6] which have gradual onsets and longer-term consequences (i.e., climate change, desertification, collapse of aquifers, and ocean acidification) and which this paper intentionally cannot address due to their complexity. In addition, although steel and composite structures also provide the best defense against man-made disasters, particularly intentional explosions, that topic will not be addressed here. Finally, the term steel will be used to refer to both steel and steel-concrete composite systems as these systems are often used together. The ability of composite structures to provide superior performance is detailed elsewhere [
7,
8].
Generically, resilience is defined as the ability to minimize the costs of a disaster, to return to a state as good as or better than the status quo ante, and to do so in the shortest feasible time [
9]. Often resiliency is depicted as some percent of the functionality of a system vs. time, with a sudden loss of functionality at the time of an event, followed by a recovery period. Figure 2 illustrates this idea along with the concept that the rate of recovery can be significantly influenced by adding or subtracting resources from the system [
10].
Other types of definitions of resiliency are also useful in the context of this paper. For example, a process-oriented definition (social sciences) defines resilience as “the capacity to adapt existing resources and skills to new situations and operating conditions.” An outcome-oriented (engineering) definition would define resilience in terms of degree of recovery, time to recovery, or extent of damage avoided. The resilience of a system’s function can be measured based on the persistence of a corresponding functional performance under uncertainty in the face of disturbances [
11].
A resilient system is one that shows: (a) reduced failure probabilities, (b) reduced consequences from failures, in terms of lives lost, damage, and negative economic and social consequences, and (c) reduced time to recovery (restoration of a specific system or set of systems to their “normal” level of functional performance) [
10]. A resilient system exhibits:
• Robustnessstrength, or the ability of elements, systems, and other measures of analysis to withstand a given level of stress or demand without suffering degradation or loss of function;
• Redundancythe extent to which elements, systems, or other measures of analysis exist that are substitutable, i.e., capable of satisfying functional requirements in the event of disruption, degradation, or loss of functionality;
• Resourcefulnessthe capacity to identify problems, establish priorities, and mobilize resources when conditions exist that threaten to disrupt some element, system, or other measure of analysis. Resourcefulness can be further conceptualized as consisting of the ability to apply material i.e., monetary, physical, technological, and informational and human resources in the process of recovery to meet established priorities and achieve goals. Its impact is shown in Fig. 2; and,
• Rapidity the capacity to meet priorities and achieve goals in a timely manner in order to contain losses, recover functionality, and avoid future disruption
In a broader sense, resiliency should be seen as a generalization of sustainability rather than the other way around. Sustainability is an attribute of dynamic, adaptive systems that are able to flourish and grow in the face of uncertainty and constant change [
12]. Achieving sustainability requires innovation, foresight, and effective partnerships among corporations, governments, and other groups. However, resiliency goes beyond this concept, as it can be seen as developing and implementing strategies to limit social, economic and political disorder in the face of natural or man-made catastrophes. Some would argue that risk management and sustainability two ends of the resiliency continuum as resiliency is a feature of systems perpetually evolve through cycles of growth, accumulation, crisis, and renewal, and often self-organize into unexpected new configurations [
12].
Sustainability of steel construction
Some facts about steel production and its uses are useful in framing the discussion about resilience of structures (World Steel Association, 2015.
www.worldsteel.org/steel-facts.html
):
• The steel industry is the second biggest industry in the world after oil and gas with an estimated global turnover of 900 billion USD.
• Average world steel use per capita increased from 150kg in 2001 to 217 kg in 2014. The distribution of the use of steel is shown in Fig. 3.
• Steel is used in every important industry; energy, construction, automotive and transportation, infrastructure, packaging and machinery. The housing and construction sector is the largest consumer of steel today, using around 50% of steel produced.
From a sustainability standpoint (World Steel Association, 2015.
www.worldsteel.org/steel-facts.html
)
• Steel is the most recycled material in the world, with over 650 mega tons recycled annually. The recovery and use of steel industry by-products has reached a worldwide material efficiency rate of 96%.
• Around 90% of water used in the steel industry is cleaned, cooled and returned to source. Most of the loss is due to evaporation.
• The energy used to produce a ton of steel has been reduced by over 60% in the past 45 years, with an additional 40% reduction in its carbon footprint in the past 25 years.
From a geopolitical standpoint, it is also important to recognize that globalization has resulted in about 35% of steel being traded internationally, and that international policies currently do not promote a level playing field to ensure that steel companies in one region are not put at a disadvantage with steelmakers from other regions or in relation to competing materials. In addition, given some of the facts stated above, governments need to recognize and embrace the importance of a strong and healthy steel industry in a sustainable economy.
The steel industry has played a key role in development, with the advent of the Bessemer process being considered as the start of the second Industrial Revolution that led to mass production of manufactured goods. The steel industry today continues to play a leading role in development, with important roles in activities ranging from alternative energy sources (wind turbines, offshore tidal and wave structures, and new nuclear power stations) to low-cost housing solutions.
Development of sustainable structural solutions
A framework for sustainability evaluation of new structural systems is shown schematically in Fig. 4. The process begins with conceptual designs, followed by experimental testing of progressively more complex and larger scale specimens, including advanced instrumentation to carefully track damage progression. The data are then used to develop robust component and entire system models. These models will serve as the basis for fragility evaluation and the subsequent resilience and sustainability assessment. This approach is structure-centric and needs to be melded into far more generic approaches that are capable of producing robust outcomes for different resiliency approaches.
Resilience decision making tools need to include state-of-the art modeling, database management, information technologies, and performance measurement methods. The models will pay close attention to the interdependencies and feedback loops between the main community sectors identified in the NIST Resilience Framework [
13]: buildings, transportation, energy, communication and water/wastewater. Taken individually, or in the aggregate, these systems are intimately linked with the economic well-being, security, and social fabric of the communities they serve. Several subsets of models are needed for various hazards effects on buildings and lifelines, and local and regional consequences. The models must be capable of incorporating interdependencies among elements in the built environment and their social and economic impacts. From these hazard and consequence models, mitigating technologies, emergency responses and recovery strategies can be developed and evaluated. The overall outputs can then be used to develop and calibrate useful resiliency metrics. There are a number of promising developments in this area with the efforts led by the new NIST Center for Risk-Based Community Resilience Planning (http://resilience.colostate.edu/). The centerpiece of the center’s effort will be the NIST-CORE—the NIST-Community Resilience Modeling Environment. Built on an open-source platform, the computer model and associated software and databases will incorporate a risk-based approach to decision-making that will enable quantitative comparisons of different resilience strategies.
In this context, there are at least two areas that have yet to reach maturity so that reliable assessments can be conducted. The first area regards the lack of comprehensive modeling tools. Achieving this goal will require the development of platforms that can utilize the large amount of data that is becoming available, as big data can provide insights not currently available through traditional tools. The models developed should include: 1) sufficiently capturing the complexity associated with built environment, social, and economic systems to develop tools for resilience measurement, valuation, and policy and decision making; 2) adequately measuring resilience attributes in meaningful metrics that lend themselves for aggregation and total economic valuation; 3) offering bases for implementation despite uncertainties; and 4) building on related existing concepts such as risk, policy and decision making, and benefit-cost ratios. In addition, any new model should consider the increasing ability of personal communication devices both to serve as data input sources and to influence disaster response and recovery through social media. This implies that while resilient structural system for different community sectors are important, structural systems per se are only one component of a resilient community. In the past it has been common to state that structural systems have performed “well” during a catastrophe; the argument in this first part of this paper is that such narrow view is counterproductive and that structural engineers need to be far more pro-active in addressing system interdependencies and analyzing potential cascading effects than in the past.
The second area regards the recovery process. Although there is a broad literature that addresses community resilience (see Refs. [
9,
14]) and damage modeling for structures has reached some maturity [
15], the prioritization of the recovery process for the built infrastructure has received scant attention. This problem was clear to the senior writer, who was in Christchurch (NZ) for a period of about three months, starting just before the 22 February 2011 event [
16], and who saw, first hand, how difficult this problem can be. Addressing this topic requires the development of:
• Damage assessment functions that compare the as-damaged, the as-built, and as-designed models, including recognition that in many instances the as-built and as-designed construction documents are unavailable. Innovative and emerging strategies to address this challenge include the creation of as-built Building Information Models (BIM) of critical infrastructure systems and the use of Augmented Reality (AR) technologies to compare the as-damaged with the as-built/as-designed models [
17]. Although the author is familiar with some isolated and successful efforts in this area, an accepted framework to address these issues has not yet emerged.
• Recovery quantification functions that commingle the magnitude of the loss in the system’s integrity, the rapidity with which the system returns to normalcy and the level of resources made available for recovery. Recovery quantification will need to recognize the interdependence between damaged infrastructure systems and the organizational approach to perform recovery efforts, damage assessment, de-commissioning, recovery design, procurement, construction, and commissioning. Innovative and emerging strategies to address this challenge include flash tracking and relational contracting strategies.
Solutions for existing structures
While technical solutions for new structures constitute the bulk of the contributions in the technical papers in this issue, it is important to recognize that the current major vulnerability comes from existing structures and lifelines. Such infrastructure suffers from poor conceptual design and detailing as well as a lack of quality control and assurance in both the construction and maintenance phases. Thus a primary goal should be to increase the resiliency of existing infrastructure to improve the social and economic aspects of recovery. A team at Georgia Tech, where the author worked at the time, began to develop some innovative solutions several years ago for fragile low-cost apartment buildings in Asia and the Americas. The criteria for these retrofits were that they:
• must result in robust and resilient buildings,
• be efficient in the use of materials, with minimal energy requirements and emissions,
• be able to achieve its goals with only minor on-site construction and disruption to existing non-structural elements,
• require little or no skilled workmanship, and the materials could be sold as a kit,
• require minimal modification to existing structural elements, and
• require little maintenance (low life-cycle cost, easy to replace).
Some of these solutions are suitable for a multi-staged incremental seismic rehabilitation strategy proposed by FEMA [
18], in which a series of discretized actions can be made to coincide with regularly scheduled building repairs and maintenance or capital improvement. Of course, variations of the devices to be described next can be installed in new construction as most of the technologies are scalable to much larger forces than the ones envisioned in the original concept development
The five proposed systems adopt a unique approach to design supplemental systems by using mostly tension-only elements. The first of these systems is shown in Fig. 5(a), and consists of a thin steel plate acting as a supplemental shear wall system [
19]. It is intended for small, low-rise steel structures in which the plate and surrounding boundary elements are installed in the middle of the bay, separate from existing columns. This geometry intends to eliminate the need to strengthen the existing columns, as these typically would have been designed only for the combined forces of gravity and wind. The system employs supplemental elements (cables) as tension-only elements to speed up the construction work and to enforce strict capacity design principles (i.e., overstrength is capped). The system achieved stable hysteretic behavior without showing major strength deterioration until large story drifts were reached (Fig. 5 (b)). A high-fidelity FE model of the system was also developed to reproduce the experimental behavior. The model well traced the test results and was used as a tool for validating the effectiveness of the proposed system geometry (Fig. 5(c)). This system could be modified for use in RC structures by providing a small metal frame anchored to the concrete beams.
The second system is a cable-based high damping bracing system (Fig. 6(a)) that consists of eccentrically connected elastic cables and a COuples REsisting Damper (CORE Damper). In this system, all cables are intended to be in tension under lateral loads so that the system dissipates energy through a bi-linear hysteresis curve [
20]. The key features of the tension-only system includes: 1) elimination of undesirable glob and local buckling in supplemental load-carrying element; 2) rational implementation of a strict capacity design (overstrength is known or capped); 3) scalability and adaptability to many bay geometries; 4) use of simple connections with rapid and adjustable installation; and 5) minimal disruptions to occupants. The CORE damper is designed to resist the torsional moment induced by coupled cable forces, and to dissipate energy through the cyclic bending of steel plates, which are easily replaceable after a severe earthquake event (Fig. 5(b)). Results of the experiment tests (Fig. 6(b)) show that the CORE damper system can sustain interstory drifts in excess of 4%, with equivalent viscous damping ratios of over 30% .
The third system to be evaluated was a recentering C-shaped articulated quadrilateral (AQ) system with shape memory alloy (SMA) cables, shown in Fig. 7(a). This device [
21] utilizes a small internal pinned rectangular frame with internal SMA cables for recentering and C-shaped frames for energy dissipation, as shown in Fig. 7(b). The design is driven by SMA’s unique ability to recover strains of up to 8% [
22] through a diffusionless phase transformation (Fig. 8). The principle behind this proposed brace is the ability to adjust the energy dissipation in a recentering hysteretic loop through the use of an AQ arrangement. SMA wire bundles were installed within the AQ and were tested alone and in parallel with C-shaped steel dampers. Though C-shaped dampers were used, a variety of options are available to provide paralleled damping. This system was also shown to have excellent energy dissipation and recentering capability, as shown in Fig. 6(b).
The fourth system considered is a conventional bracing system that utilizes a shape memory alloy (SMA) helical coil tension/compression damper, as shown in Fig. 9(a). The prototype is fabricated using a solid NiTi spring as the active element [
23]. The prototype shaft is approximately 50 cm in length, 6.4 cm in diameter, and constructed of stainless steel. The SMA spring was constructed of a 2.54 cm diameter rod, set in the form of a helical spring. As shown in Fig. 9(b), the device has excellent recentering capability, and damping values in the range of 5-10% equivalent viscous damping. Peak force and stroke values can be modified by varying the properties of the SMA coil. Detailed analytical studies of multi-story buildings with similar bracing systems subjected to large ground motions have shown that the SMA bracing systems reduce the peak interstory drifts by an average of 75%, compared with similar steel bracing systems, and reduce peak residual drifts by over 90%.
The fifth system [
24], similarly to the fourth one, is intended to replace a conventional brace with a hybrid device consisting of two mild steel tube struts for energy dissipation that are attached outside of a set of two freely-floating high-strength steel tubes filled with SMA wires for re-centering capability (Fig. 10(a)). Substantial improvements in energy absorption, re-centering, and cost as compared to extant damper concepts are possible with this hybrid device (Fig. 10(b)). To maximize both the energy absorption and re-centering capacities, the reverse transformation yielding force provided by the SMA wires has to be greater than the yielding force developed by the energy-absorbing struts. For the maximization of energy absorption, the potential for buckling in the struts has to be minimized when they are subjected to large compression cycles.
The five systems briefly described above meet or exceed all the criteria described initially for improving resiliency, and they are but a small subset of what can be done today through combinations of innovation geometrical configurations and new materials for both existing and new structures. All the systems intend to improve not only the collapse capacity but also to minimize residual deformations though some sort of self-centering capability. These solutions can be implemented, with appropriate modifications, not only in buildings but in all types of infrastructure.
Solutions for new structures
The robustness of steel structures designed to modern standards was strikingly evident after the 2011 Christchurch Earthquake [
25]. While reinforced concrete structures in general met the life safety standard, extremely few remained serviceable after the earthquake. The vast majority of the large RC buildings have been deconstructed. Steel structures, on the contrary, performed exceedingly well and observed damage was easily repairable. Many of them remained serviceable and were used as part of the recovery effort.
From the performance standpoint, limiting residual deformations is a convenient but indirect index to quantify both damage to structural and non-structural elements and direct and indirect economic losses. Limiting residual deformations can be accomplished by implementing self-centering concepts [
26], ranging from posttensioning frames [
27,
28] to modifications of the horizontal (Fig. 11) or vertical systems (Fig. 12). The system in Fig. 10 [
29] consists of a truss configuration with pre-compressed concentric tubes that allowed for gap openings as the truss undergoes cyclic lateral deformations. The issue of deformation incompatibility is solved by moving the gap openings within the truss at the bottom chord. This prevents gap openings from interacting with the diaphragm directly or through bay elongation. The energy dissipation was permitted through the use of highly ductile and resilient steel butterfly fuse plates to create a moment resisting frame with superior seismic performance. Similar goals can be achieved, as shown in Fig. 12, by using fuse plates as energy dissipaters in foundation rocking systems and post-tensioning for recentering [
30]. Many such systems have been proposed and studied over the last decade and are beginning to find practical applications [
31].
This theme issue of Frontiers of Structural and Civil Engineering contains six additional papers that describe similar efforts, including:
1) The paper by Dong, et al. describes a novel lateral force resisting system consisting of nonlinear horizontal viscous dampers and associated diagonal bracing. The system is designed to control the lateral drift demands and was tested using real-time hybrid simulation on a 0.6 scale three story structure. The building demonstrated superior performance under DBE and MCE ground motions for cases corresponding to 100%, 75% and 60% of base shear design strength required by ASCE 7-10 for a structure without dampers.
2) The paper by Maurya and Eatherton describes the background and design of a shop-fabricated self-centering beam and details the results of an experimental program on five specimens at 0.67 scale. The specimens underwent 5-6% story drift without any observable damage to the self-centering beam and columns. The analytical strength equations developed for the self-centering beams predicted the moment capacity satisfactorily, with a mean difference of 6% between experimental and predicted capacities.
3) The paper by Yang and Li describes a buckling restrained knee braced truss moment frame that utilizes the advantages of long-span trusses and dedicated structural fuses for seismic applications. The system uses buckling restrained braces as the designated structural fuses to dissipate energy, and was developed based on performance-based plastic design principles. The study shows this system can effectively control the structural and non-structural component damage and minimize the repair costs of the structure under different ranges of earthquake shaking intensities.
4) The paper by Chou describes both energy dissipation and self-centering properties of a dual-core self-centering brace. Tests of individual components as well as of the first story of a three-story dual-core self-centering braced prototype frame indicated good behavior, with a maximum residual drift of the DC-SCBF caused by beam local buckling of 0.5% after 2% drift cycles.
5) The paper by Clayton et al. describes a self-centering steel plate shear wall system, consisting of thin steel infill panels that serve as the primary lateral load-resisting and energy dissipating elements, while post-tensioned beam-to-column connections provide the system with recentering capabilities. Quasi-static subassembly tests, quasi-static and shake table tests of scaled three-story specimens, and pseudo-dynamic tests of two full-scale two-story SC-SPSWs were carried out. This paper discusses innovative PT connection and web plate designs that were investigated to improve constructability, resilience, and seismic performance and that can be applied to other self-centering and steel plate shear wall systems.
6) Finally the paper by Silva et al. describes the development of concrete filled tube columns infilled with rubberized concrete (RuC), which included a total of 36 specimens utilizing a novel testing setup, aimed at reducing both the preparation time and cost of the test specimens. The test results show that the bending behavior of CFST elements is highly dependent on the steel tube and that the seismic design of composite moment-resisting frames with these columns according to Eurocode 8 not only leads to lighter design solutions but also to enhanced seismic performance and resilience in comparison.
A Recentering connection to circular concrete-filled tubes
A more radical and unconventional approach is being studied by the author and his students [
32,
33], as well as others [
34] by modifying conventional practice through balancing deformations between a conventional beam yielding mechanism, including the reduced beam section (RBS) concept, and a partially restrained recentering connection. A 3D view of a biaxial implementation of the proposed connection is shown in Fig. 13, while the details of a 2D connection are shown in Fig. 14. The connection consists of:
• A circular concrete filled tube (CCFT) column made from either HSS sections or tubes of larger dimensions and higher strengths, such as those used in bridges or offshore structures.
• Conventional beams with reduced beam sections (RBS) that limit the force input to the connection while maintaining the frame stiffness.
• A stiff extended end plate connection with stiffeners to transfer all girder forces into the connections through threaded rods. End plates are economical and robust connections [
35]
• The end plates will bear on a rectangular tube, which in turn will be connected to the circular column by fillet-welded internal diaphragms and external cover; the space between the tubes will be filled by a high strength expansive grout.
• A series of large diameter rods composed of both shape memory alloys (SMA) and mild steel rods. The design of the mild steel rods minimizes inelastic deformations during moderate seismic events and provides additional strength at ultimate while the design of the SMA rods [
36] provides re-centering after large events. The forces need to be carefully apportioned between the two types of rods and prestressing of both types of rods may be needed to maximize performance.
While this connection may appear complex, it compensates for initial costs through shop prefabrication, ease of erection, self-centering characteristics and higher robustness.
The behavior of this connection is not intuitive, as shown by the moment-rotation characteristics in Figs. 16 and 17 when subjected to the two load histories shown in Fig. 15. The coordinates of the anchor points in in Figs. 16 and 17 are fixed for a specific connection because they represent several key points calculated based on the design procedure:
• Point A stands for the maximum capacity of the connection at rotation of about 0.05rad, which is the rotation of the connection when the maximum moment from the connecting beam is reached as determined from the proposed design procedure.
• Point B stands for the maximum capacity of the connection at a rotation of about 0.045rad, which is the rotation of the connection when the maximum strain of some SMA rods reaches 5%.
• Point C stands for the flexural capacity of the connection supplied only by the SMA rods at a rotation when the maximum strain in the most highly loaded SMA rods reaches 5%.
• Point D stands for the flexural capacity of the connection from the pretension of only the SMA rods. The corresponding rotation of the connection is nearly zero.
• Point E stands for the flexural capacity of the connection from the pretension of both the SMA and the steel rods. The corresponding rotation is also nearly zero.
• Point F stands for the flexural capacity of the connection from only the SMA rods at rotation of 0.0075rad, which is the rotation of the connection when the SMA rods enter the phase transformation stage.
Three backbone curves characterize strength and stiffness behavior of these connections as shown in Figs. 18 and 19. Insofar as strength, Fig. 18 shows (1) an upper backbone curve that represents the capacity of both the steel and SMA rods; (2) a middle backbone curve that represents the upper shelf of the SMA rod strength (Fig. 8); and (3) a lower backbone curve that represents the lower shelf strength of the SMA rods. The difference in moment capacity at a specific connection rotation between the middle and the upper backbone curves is the moment capacity supplied by the steel rods under that specific deformation. Similarly, the difference in moment capacity at a specific connection rotation between the middle and lower backbone curves reflects the different strength of the SMA rods under that specific deformation on the upper and lower backbone curves of the ‘Flag-Shape’ stress strain curve (Fig. 8).
The differences in behavior between the first and third quadrants in Fig. 16 is due to the symmetry of the load history, which leads to some yielding of the steel rods in the positive direction but only reaching the yield strength in the negative direction, resulting in a series of spikes or “pins” in the curve. Similarly, the differences in load histories lead to the formation, shown in Fig. 17, of alternating “teeth” between the middle and upper backbone curves in the first and third quadrants. By alternating it is meant that, if superimposed, the “teeth” in the negative and positive direction in Fig. 17 would result is a similar pattern as in the positive direction of Fig. 16. The respective stiffnesses, described in Fig. 19, need careful monitoring when modeling as small but perceptible deviations from straight lines occur as the unloading from the upper to the middle backbone curve occur. For the case of a real seismic excitation, where the displacement history is random, the curves are more complex.
A modified version of the connection described above labeled SC-Gaps, in which initial gaps were introduced in the conventional steel rods to improve the global behavior was used in parametric frame studies of a public library building. The library has five stories above the grade and a one story basement designed for a Site Class D soil (Vs>280 m/s) in a high seismic zone. The plan dimensions of the building are approximately 36.5m. in the N-S direction, and 49 m. in the E-W direction. The structural system is intended to work within a grid spacing of 9.7 m by 7.3 m with story heights of 4.2 m.
Figure 20 shows a typical comparison of the behavior of a frame with the innovative connection vs. one with conventional welded connections under one of the 44 ground motions used. All members are for the E-W frame (9.7 m) and are the same size for both the SC-Gaps and conventional frames; the only difference is in the connection themselves A very large difference in residual drift is evident, and this result was consistent through all ground motions as shown in Fig. 21.
Closure
This brief survey paper on resiliency intends to demonstrate that steel structures provide the required behavior to promote community resiliency through sustainable and low-risk solutions to the design of infrastructure. A number of innovative devices applicable to both new and existing structures were reviewed. An innovative partially-restrained connection was introduced and its behavior described.
Higher Education Press and Springer-Verlag Berlin Heidelberg
PDF
(1054KB)
