Structural Engineering Research Frontiers

Explicit integration procedures have been widely adapted and applied to hybrid simulations of the seismic response of structures due to their simplicity. However, these procedures are conditionally stable and require small time steps for stiff systems, or systems involving high-frequency modes. In this study, a combined implicit or explicit integrator is proposed to solve the equation of motion for hybrid simulations. In this method, command displacement predictions are determined to load the experimental substructures beyond the current step using explicit kinematics expressions. The displacement and force measurements are then used to iteratively correct the explicit displacement prediction within the numerical model until an implicit formulation is satisfied. An explicit solution is accepted only in time steps for which the iterative correction of displacement fails to convergence after maximum number of iterations. This procedure limits communication between the experimental substructures and the numerical integrator, thus it is well suited for networked applications of hybrid simulation. Numerical and experimental simulations demonstrate the effectiveness of this numerical integration scheme, especially in utilization of longer time steps, prevention of excitation of higher modes, and testing of stiff systems.

Real-time hybrid pseudodynamic (PSD) testing of a 3-story steel moment resisting frame (MRF) equipped with elastomeric dampers at the first floor is presented. The components of the servo-hydraulic integrated control system of the NEES Real-time Multi-directional (RTMD) Earthquake Simulation Facility at Lehigh University are described, along with the analytical substructure modeling, experimental substructure properties and the experimental setup. The stability boundary with respect to both the analytical and experimental substructure restoring force delay is evaluated. In order to minimize the delay in the measured restoring force from the experimental substructure, a velocity feed forward component is introduced into the servo-control loop and its contribution is demonstrated both analytically and experimentally. Selected experimental results of real-time hybrid PSD tests are presented.

Hybrid simulation is a versatile, powerful and economically viable experimental method. In particular, through the use of numerical partitioning concepts, it should be possible to test portions of a large structure in one of more laboratories, and to simulate the remainder of the structure/foundation system numerically. Considerable efforts have been expended to develop such capabilities, but their adoption has been hampered by the absence of a common framework for conducting hybrid simulation across various laboratories and computing environments. Therefore, based on object-oriented software design methodologies, a set of interrelated software classes is described herein, which in their unity form a network-enabled framework for integrating experimental testing with standard structural analysis frameworks. In this paper, the open-source Framework for Experimental Setup and Control (OpenFresco) is used in combination with the object-oriented finite element software framework OpenSees (Open System for Earthquake Engineering Simulation) and the graphical user interface OpenSees Navigator. This enables an unlimited number of users to connect multiple laboratories throughout the world, to define specialized experimental setups, to implement advanced integration operators and to utilize high performance control methodologies, in a highly modular and scalable fashion. By utilizing both the object-oriented experimental and finite element software frameworks, response features that are frequently encountered in large structural systems (e.g., geometric nonlinearities, three-dimensional effects, multiple support excitation and soil-structure interaction) can be investigated by incorporating them into the analytical model. One of the most advantageous applications to incorporate geometric nonlinearities into the analytical model is in large displacement tests where a structure is tested until collapse. Contrary to shaking table testing, no large physical masses have to be present in the experimental part of a hybrid simulation. Another example presented illustrates the ability to account for multiple support excitations in long span structures. By using a standardized framework for communication, simulation and control, such tests can be easily set up in various laboratories to test large structures not possible by other means.

The development and implementation of a novel real time seismic response simulation method for large to full-scale structures, which involve the combined use of shake tables, dynamic actuators and computational engines is presented. With this technique, it is possible to test a substructure at full scale, or close to full scale, while the rest of the structure is simulated by computers. At the end of the synchronized analysis and testing it is possible to obtain realistic response of the entire system. The method was defined as Real Time Dynamic Hybrid Simulation (RTDHS). The structure to be simulated is divided into one or more physical-experimental and computational substructures. Interface forces between the physical and computational substructures are imposed by actuators and the resulting displacements and velocities are fed back to the computational engine. The dynamic excitation, such as an earthquake ground motion, is applied to the physical substructure by shake tables. The unique aspect of the above hybrid simulation method is the force-based substructuring. Since the shake tables induce inertia forces in the experimental substructures, the actuators at the interfaces have to be operated in dynamic force control as well. The simulation method presented herein is more versatile than the existing seismic simulation methods. The substructuring strategy related to RTDHS is presented focusing on the structural model condensation for real time simulation purpose and on interface force calculation. The numerical integration algorithm associated with the computational substructures is discussed along with the implementation of the computational engine. Issues related to the implementation of RTDHS are identified including force-controlled strategy in dynamic actuators and time delay compensations. A unique compensation controller is designed and experimentally verified. An example of a real-time hybrid test implementation, and results from this experiment are also presented.

This paper discusses the on-going shake table testing program on a full-scale two-story woodframe townhouse building conducted within the NSF/NEES-funded NEESWood Project. The test building represents the worlds largest woodframe structure tested on a shake table. The size and weight of the test structure required the simultaneous use of the two tri-axial shake tables in the Structural Engineering and Earthquake Simulation Laboratory at the University at Buffalo. The testing program is focusing on the various construction elements that may have significant influence on the seismic response of woodframe buildings and that should be considered in performance-based seismic design. This paper focuses on the effects of gypsum wallboard finishes. The results obtained so far indicate that the influence of gypsum wallboard finishes applied to the interior surfaces of structural (load bearing) walls substantially improved the seismic response of the test building. On the other hand, the same gypsum wallboard wall finishes applied to interior partition walls had no beneficial effect on the seismic response of the test building because of the lack of structural connections between those partition walls and adjacent floor and roof diaphragms.

A collaborative research project is underway at the University of Minnesota Multi-Axial Subassemblage Testing (MAST) Laboratory regarding the behavior of structural wall systems. The investigation included tests and numerical simulations of three rectangular reinforced concrete wall systems to investigate the effect of continuous, spliced, and mechanically connected longitudinal reinforcement at the wall-foundation interface. It was anticipated that the different reinforcement details would have an effect on the plastic hinge length, local strain demands, and consequent flexural (deformation) response including the damage state of the wall systems. The walls were instrumented to investigate the overall behavior and to isolate the individual deformation components attributed to flexure and shear. Understanding the sources of deformation and correlation of the results with models that can be used to predict the behavior is important for development of performance-based seismic design procedures. The project was funded by the National Science Foundation (NSF) grants CMS0324504 and CMS0324559.

A quarter scale model of a two-span bridge was tested using the NEES multiple shake table system at the University of Nevada, Reno. The project was funded through an NSF-NEES demonstration grant. The prototype was designed using the provisions of the National Cooperative Highway Research Program document 12-49 for seismic design of highway bridges. The input shake table motions included motions corresponding to the expected and rare earthquakes with probabilities of exceedance of 50% and 3% in 75 years, respectively. The bridge was designed for "life safety", the minimum seismic design requirement. Test results demonstrated that the model met the performance objectives for both earthquakes in terms of damage, rotational capacity, and ductility capacity.

The George E. Brown, Jr. Network for Earthquake Engineering Simulation (NEES), operational since October 2004, is intended to transform the nation's ability to carry out earthquake engineering research. NEES consists of fifteen shared-use equipment sites located throughout the U.S. that include a variety of state-of-the-art test equipment including large-scale hybrid test facilities, shaking tables, geotechnical centrifuges, a tsunami wave basin, as well as field equipment, and an information technology (IT) infrastructure linking these sites over high-speed internet. In 2004, the National Research Council (NRC) developed a report, at the request of the NSF, recognizing the potential of NEES to accelerate the pace of earthquake hazard mitigation and identifying research activities NEES should pursue (NRC, 2004). This report specifically identifies research activities involving tests at multiple equipment sites as a major focus of the geographically distributed NEES collaboratory. In addition to collaboration and data sharing between sites, NEES was charged to develop distributed physical and numerical simulation capabilities to conduct more complex experiments and simulation tests simultaneously at multiple equipment sites. This ability will accelerate the pace of earthquake hazard mitigation. Over the past 2 years of operation, the NEES community has focused on multi-site pseudo-dynamic testing including the Multi-site Online Simulation Test (MOST) and ensuing bench-scale mini-MOST demonstrations, and pre-NEESR zipper frames project (Leon, et al., 2004). Multi-site NEES activities are limited to pseudo-dynamic testing, that is, tests run at a much slower rate than real-time. The major challenge with distributed, or multi-site, testing is accommodating the large communication time delays present in sending data over large distances and over the internet. Local controllers and hybrid algorithms run at a rate of around 1000 Hz (e.g. the University of Colorado at Boulder NEES Fast Hybrid Test facility runs at a rate of 1024 Hz). 1000 Hz corresponds to a time step of 1 millisecond (msec). A larger time step, using the current methods and algorithms may result in less accurate simulation tests. it is generally accepted that communications over the internet would be limited to 75 Hz, over a 500 km distance. Assuming the test rate of 75 Hz is mainly a result of communications delays from both sending and receiving data, the communications time delay in one direction is on the order of 6 msec. Because of this limit in the rate of multi-site experiments, testing in real-time (which would allow for testing rate dependent physical components) can not be accomplished with the current methods and algorithms. NEES equipment sites such as the fast-hybrid testing facilities, shake tables, centrifuges, and tsunami wave basin were specifically built for fast and/or real-time testing. In order to demonstrate a potential method to achieve real-time multi-site testing, a simplified experiment is devised and implemented to simulate multi-site testing in real-time. This simplified experiment will seek to experimentally verify semiactive structural control of a seismically excited building. This paper proposes a system to conduct real-time multi-site experiments. A bench-scale experiment of a two-story building with semiactive control device located between the ground and first story is presented to illustrate this method, employing hybrid testing of a physical semiactive MR damper while simulating in real-time the seismic response of a two-story building model. This paper first proposes a system architecture that applies techniques from haptics to overcome simulation to experiment interface and delays anticipated in multi-site testing. Next, the test setup for experimental verification is described. Results from these tests are presented. Conclusions are drawn and future studies are identified.

Pre-analytical research for the six-story reinforced concrete building tested at E-Defense has been done before the test was finished at January, 2006. Our pre-analytical results predicted either the base shear force or the final collapse type of test results with good accuracy by original JMA Kobe wave excitation. In this study, analyses are done by six motion components measured from test including rotations in three-dimensions. Usually we use only one panel model for one story shear wall. But from test results, we find the damage of shear wall is severely different from bottom to top at first story. In order to present the different damage at bottom and top of the first story shear wall, two analysis cases are done by the first story shear wall is modeled by one panel model and divided into two panel models. The main difference between two analytical methods shows on the evaluation of shear deformation.

Guidelines have been developed in the United States to assist in the design of buildings resistant to progressive collapse. These guidelines outline several methods for incorporating the dynamic amplification on load redistribution that occurs when there is sudden member failure in a structure. Analytical studies have been performed to evaluate the consistency of these guidelines, but very little testing has been conducted to verify the guidelines and analytical work performed to date. This paper describes a field test in which one column of a two-story reinforced concrete frame building was removed by explosives while the resulting dynamic changes in axial load of surrounding columns were monitored. The data collected during this test is being used to assess the effects of dynamic amplification on gravity load redistribution. Observations from the test suggest that the building frame remained linear elastic. The peak axial force increases recorded in the columns were more than twice the corresponding steady-state increase, exceeding the limit expected by linear elastic theory. The high peak-to-steady-state force ratios observed are likely attributable to the magnifying effect of the pressure pulse caused by the explosion. A single degree of freedom model has been developed that provides a good approximation of the axial force response of a column adjacent to the column explosively removed in the test. This model has been used to remove the part of the recorded response caused by the pressure pulse and thereby estimate the response solely due to gravity load redistribution. Refinement of this model is ongoing. In addition to being applicable for blast loads, this research program is also relevant for progressive collapse analysis involving earthquake loads. Since column axial-load failures resulting from seismic forces occur at a slower rate when compared with blast failures, the dynamic effects resulting from the former are less severe. Thus, it will be reasonable to implement the findings of this research program based on blast loads as an upper bound when addressing the influence of dynamic amplification on the gravity load redistribution that occurs during the collapse of reinforced concrete frames in earthquakes.

In order to simulate the collapse process of a six-story full-scale wall-frame building tested at E-Defense in January 2006, analytical studies were carried out before and after test. In the preliminary analysis, the effects of analytical model characteristics, shear-span to depth ratio of short column and torsional response on seismic behavior of a specimen were investigated. And after test, numerical simulation of the experimental response was conducted by modeling the behavior of each frame with different kinds of models, which are characterized by the capability of simulating strength degradation. Simulated results showed that the observed strength-degrading feature in post-peak region could be reproduced from the analytical models employed in shear critical members such as short column and shear wall.

Response of reinforced concrete wall-frame structures subjected to earthquake motions was investigated by testing a full-scale six-story specimen on the world's largest tri-axial shake table "E-Defense". The specimen was designed based on the code of design and practice in 1970', and columns and walls were defined as flexural yielding type based on current code calculation. The specimen was subjected to series of earthquake motions and finally collapsed due to shear failure in short columns and the structural wall at 1^st story during a strong earthquake motion.

Collapse experiments are found very helpful in facilitating collapse analysis of new buildings and identification of older buildings that are at high risk of structural collapse during severe earthquake events. Experimental observations from dynamic global collapse of a single story 3-bay RC frame are presented in this paper. The frame is composed of shear-critical columns and ductile columns to allow for load redistribution. A near-fault record from the September 21 (local time) 1999 Chi-Chi Taiwan earthquake was employed. Preliminary investigation shows that existing empirical equations are able to predict hysteretic backbone until the point of structural collapse with satisfaction.

This paper presents the results of a study to simulate progressive collapse in a structural system with seismically-deficient components. The range of investigated seismic deficiencies includes inadequate transverse reinforcement, inadequately-developed lap splices, and insufficient lateral confinement leading to buckling in longitudinal bars and/or rupture in transverse ties. The progression of damage into a deficient component is tracked using component models calibrated to trigger an element removal algorithm upon the loss of load-carrying capacity. The component models include a strain-compatible confined fiber section model, confinement-sensitive uniaxial material models for concrete, lap-splices, buckling-enabled longitudinal reinforcement, and axially-coupled shear springs. These component models are applied to the finite element model of a structural system to investigate the effect of older construction details on its progressive collapse potential. The structural system consists of three one-story reinforced concrete (RC) frames linked by a RC slab and transverse spandrel beams. The middle frame is infilled by an unreinforced masonry wall. Intentional deficiencies are introduced to the system to render it seismically vulnerable. The analysis procedure is carried out successfully and the element failure sequence and mode are identified until complete collapse. The simulated response to several ground motion levels highlights the path-dependency of the progressive collapse problem and the necessity of applying a probabilistic approach to the assessment of collapse potential.

Full-scale shake table tests were planed and conducted at E-Defense from September to November 2006 in the second phase of the Dai-Dai-Toku project. Two three-story specimens on flexible foundation, the bare specimen and the retrofit specimen, were tested with the following specific objectives: (1) the simulation of progressive collapse of existing school buildings, (2) verification of strengthening effect by the attached steel frames, and (3) soil-structure interaction with flexible foundation. Backgrouds of the test plan and the details of the specimens are reported in another paper, while analyses and mesured responses are reported in this paper. For the planning of the test specimens, preliminary nonlinear pushover analysis and dynamic analysis were carried out. The bare specimen was supposed to fail in shear before/after flexural yielding of the short columns. The effect of the dislocation at the flexible base and the brittle shear failure of the columns on the failure modes were considered in the analysis. An overall mechanism of rocking was to be observed in the retrofit specimen with the steel braces and seismic slits. In this paper, the preliminry analyses of the two specimens and the observed behavior of the two specimens in the shake table test are presented mainly focused on the failure mechanisms of the structure considering the base conditions. The results of the static test to identify the base friction coefficient are also reported.

E-Defense, the world largest three-dimensional earthquake simulator, has been operated and available for shake table tests since 2005. A five-year national project on seismic safety of urban areas, referred to as DaiDaiToku project, has started from 2002 in Japan. As a part of the project, seismic performance of existing reinforced concrete (RC) buildings were investigated through the full-scale shake table tests at E-Defense in 2005 and 2006. The second phase tests were conducted for two three-story school buildings with flexible foundation from September to November 2006. One was a bare RC specimen simulating an old and non-ductile school building in Japan. The other was a retrofit specimen, which was constructed in the same design and strengthened with attached steel braces. The research backgrounds, the plan and objective of the test, the design of the specimens and test procedure are presented in this paper, while detailed results with analyses are described in another paper.

Wide-ranging participation from the US and Japan is being encouraged in the NEES/E-Defense Collaboration Research Program utilizing the E-Defense shaking table facility. The facility, whose construction was completed in early 2005, is the largest earthquake simulator capable of subjecting full-scale structures to the strongest earthquakes recorded in the world. Relatively short term collaboration has already taken place involving geo-structures, wooden buildings, and reinforced concrete buildings. Currently, more emphasis is made on longer term collaboration out to 2010, involving steel buildings and bridges. Figure 1 shows overall organizations for the steel building project led by the Japan team. The steel building project focuses on moment-resisting frames, innovative methods for new or existing buildings, nonstructural elements, and protective systems. Accordingly, the Japan team has four working groups (WG's) shown in Figure 1. This paper discusses the work of Damper and Isolation Systems WG focusing on protective systems, such as passive control systems and base isolation systems for the steel buildings. Three companion papers by Suita et al., Tada et al., and Takeuchi et al. discuss the studies by the other WG's.

A major research project on steel buildings utilizing the E-Defense three-dimensional shake table facility is underway in Japan. It involves several studies on moment frames, innovative methods for new or existing buildings, protective systems, and nonstructural elements. As one of the studies, a challenging seismic experiment to simulate collapse of a full-scale bulding comprising the moment frame will be conducted in September 2007. Collapse is defined as the state where the structure loses its ability to sustain gravity loads. To-date, only a few full-scale tests have been conducted for the multi-story steel moment frames, and none of those simulated collapse. Two decades ago, a 6-story moment frame was pseudo-dynamically tested by static actuators simulating severe earthquakes. Recently, a 3-story moment frame including exterior walls was tested up to an overall drift angle of 1/15 rad, by static loads to reflect the effects of the ground motion that is greater than considered in the current seismic codes. These full-scale tests employed quasi-static loading, since a dynamic loading facility capable of much more realistic simulations was not available. The E-Defense shake table which commenced operation in 2005 is capable of subjecting a full-scale structure to the strongest ground motion recorded in the world. With a 12,000 kN specimen placed on 15m x 20m table, the table can produce a maximum velocity of ±2.0m/s and a maximum displacement of ±1.0m. With the E-Defense shake table, we have the first opportunity ever to simulate the behavior of full-scale structures up to collapse, and to obtain detailed response data. This paper outlines the test plan of collapse simulation, as well as technical issues being addressed during the process of test preparation.

The world's largest earthquake simulator, the E-Defense shake table facility is being utilized for a major research project on steel buildings in Japan. The project involves several studies on moment frames, innovative methods for new or existing buildings, protective systems, and nonstructural elements. As one of such studies, a challenging full-scale experiment to simulate collapse of a 4-story building comprising moment frames is scheduled for September 2007. Regarding this experiment, extensive analytical studies have been conducted by the writers, in order to investigate important key issues such as: possible scenario of collapse and corresponding design of specimen, selection of ground motions for shaking, member modeling techniques for accurate simulation of structure responses, and so on. The results from the analyses will be discussed in this paper. The analytical methods used will be explained next.

The E-Defense shake table facility, the world's largest earthquake simulator, is being utilized for a major research project on steel buildings in Japan. The project involves several studies on moment frames, innovative methods for new or existing buildings, protective systems, and nonstructural elements. E-Defense can shake a large specimen with a complete three-dimensional configuration. However, since the cost of such a specimen can be very high, testing a part of the specimen such as the plane frame of a building would be much more economical and can still produce meaningful data. Such test methods would also enable parametric studies requiring multiple specimens. Accordingly, the writers are currently constructing a so-called "test bed" having a multipurpose inertial mass system. The test bed will be utilized for the aforementioned study on innovative methods, involving researchers from both US and Japan, as a part of the NEES/E-Defense Collaboration Research Program. The test bed needs to accommodate various specimens, and have the following components: 1) Inertial mass connected to the specimen at each story level, with its self-weight supported, thereby inducing a horizontal inertial force into the specimens. 2) Inter-story stiffness element at each story level, having the capability of setting the stiffness to zero or other values, thereby adding desired horizontal stiffness to the specimen.

In the United States adoption of formal building codes that rigorously prescribe means and methods of design and construction of buildings is a relatively modern concept, originating in the late 19^th century, and taking most of the 20^th century to become truly ubiquitous. The absence of building code adoption in a locality, of course, did not prevent the design and development of buildings. Particularly in the era before the early twentieth century, if a building code was not available to prescribe the way a design was to be performed, engineers designed structures based on their personal knowledge of materials and systems behavior and performance. Unconstrained by prescriptive requirements, engineers designed truly remarkable and unique structures such as the 10-story Home Insurance Company building, the first official "steel skyscraper" constructed in Chicago in 1885; the Brooklyn Bridge (1883); the Washington Monument (1844), and many other monumental structures that still provide outstanding service to this day.

A primary goal of seismic design requirements of building codes is to protect the life safety of building inhabitants during extreme earthquakes. Fundamentally, this requires that the likelihood of structural collapse be at an acceptably low level. However, building codes and standards are largely empirical in nature, such that the implied collapse safety is not known. The primary objective of this study is to quantitatively predict the collapse safety of twelve modern reinforced concrete (RC) special moment frame (SMF) buildings designed according to the governing provisions of the 2003 IBC [ICC 2003], ASCE7-02 [ASCE 2002], and the ACI 318-02 [ACI 2002]. In particular, this study examines the effects of building height on collapse performance and the related effects of building height on static overstrength, building deformation capacity (as measured by roof and interstory drifts at collapse), and the collapse mechanism. The secondary objective is to examine how changes to the required minimum base shear introduced in the 2005 edition of ASCE7 [ASCE 2005] affect collapse safety. These predictions of collapse safety are made using tools and methods that have been developed by the authors and other researchers in the Pacific Earthquake Engineering Research (PEER) Center, including careful consideration of the important uncertainties associated with ground motion variability and uncertainty in nonlinear structural response. The twelve buildings examined in this paper are part of a more extensive study, which provides further details on the assessment methods and finding for a broader set of buildings [Haselton, 2006].

A second-generation performance-based earthquake engineering (PBEE-2) procedure has been developed by the Pacific Earthquake Engineering Research (PEER) Center and others that estimates the probabilistic future seismic performance of buildings and bridges in terms of system-level decision variables such as repair cost, casualties, and loss of use (dollars, deaths and downtime). The Applied Technology Council has undertaken to transfer the methodology to professional practice. The methodology involves four stages: hazard analysis, structural analysis, damage analysis, and loss analysis. This paper addresses the damage analysis, whose inputs are demand parameters calculated in a suite of multiple-degree-of-freedom nonlinear dynamic analyses, and whose output is the damage state of each damageable structural and nonstructural component in the facility. The analysis uses fragility functions, which in this context give the probability that a component, element or system will be damaged to a given or more severe damage state as a function of a single predictive demand parameter such as story drift or floor acceleration. One such fragility function is required for each component type and damage state. Many building-component fragility functions have been created in the past, but no comprehensive set of procedures exists on how to create them. This paper summarizes such a standard developed for ATC-58. See Porter et al. for more detail, examples, commentary, and alternative approaches.

This paper focuses on providing a practical approach for Performance-Based Design (PBD). The process for probabilistic Performance-Based Assessment (PBA), as developed recently in various organizations (e.g., Pacific Earthquake Engineering Research Center), implies that, given a building and its location, one can rigorously calculate losses (e.g., expected monetary loss at a specified hazard level or mean annual loss). This process is mathematically correct and powerful, but its implementation in practice is in many cases too complex and impractical. Also, it is neither practical nor efficient to use a rigorous probabilistic approach for conceptual Performance-Based Design (PBD), in which the building first needs to be created before it can be evaluated. In the simplified PBD approach proposed in this paper, performance objectives are defined by placing limits on direct (monetary) loss, downtime loss, and life loss (or tolerable probability of collapse). The proposed design process incorporates different performance objectives up front, before the structural system is created, and assists engineers in making informed decisions on the choice of an effective structural system and its stiffness (period), base shear strength, and other important global structural parameters. The tools needed to implement this design process are (1) hazard curves for a specific ground motion intensity measure, (2) mean loss curves for structural and nonstructural subsystems, (3) structural response curves that relate, for different structural systems, a ground motion intensity measure to the engineering demand parameter (e.g., interstory drift or floor acceleration) on which the subsystem loss depends, and (4) collapse fragility curves. The proposed process is illustrated in an example to demonstrate its practicality.

The tenets of performance-based design have been around for decades, yet are just recently being applied to the design of high-rise superstructures. Successful projects using a ductile core wall structural system, using a performance-based design methodology, have been implemented by Magnusson Klemencic Associates and can be seen in completed structures around the world. However, provisions often arbitrarily placed in the current prescriptive Building Code (regarding height limits and building response, for example), make it difficult for structural engineers to apply rational engineering methodologies in design of performance-based structures. Instead, engineers must incorporate less-than-ideal design practices, such as requiring redundant dual frame systems on buildings more than 240 feet or allowing construction compromises in the impractical regulations governing diagonal steel reinforcing of coupling beams. Additionally, the lack of peer review process standards often pit two well-intended opinions against each other, with the building design usually changed per the reviewer's wishes and for the sake of being expeditious. To improve the design and construction process for performance-based tall buildings, among other things, appropriate interpretations of the Building Code should be permitted, and even required, and the scope and authority of peer reviews should be standardized.

NEESit is a service-focused organization that operates and supports the extensive IT infrastructure of the Network for Earthquake Engineering and Simulation (NEES) Collaboratory. The mission of NEESit is to provide software and services that link earthquake researchers with leading edge computing resources and research equipment, allowing collaborative teams at equipment sites or from remote locations to plan, perform, and publish their experiments. The software tools are classified into four categories: data management software for organizing and sharing data, telepresence tools for enabling remote participation in experiment planning and execution, visualization capabilities for viewing historical sensor data and synchronized video images, collaboration services for joint research effots and simulation software for computational modeling. The infrastructure services are classified into three categories: user support, supercomputing allocations, and a community code repository. NEESit has also pursued education and outreach strategies to promote learning and communication amongst the IT and earthquake engineering communities. This paper discusses the IT software tools and infrastructure services, which integrate custom-built software with commercial, off-the-shelf components.

The National Science Foundation's George E. Brown, Jr. Network for Earthquake Engineering Simulation provides a singular opportunity to advance dramatically the knowledge needed for engineers to reduce the risk of catastrophic earthquakes. The physical simulation of systems in the laboratories at the NEES equipment sites provide unparalleled opportunities for researchers to investigate the behavior of components and systems. Nevertheless, experiments are a simulation because real in situ systems have to be approximated in scale, boundary conditions, and loading. Even full-scale shake table testing often neglects effects such as soil structure interaction. A second type of simulation is based on computation and it involves creating a mathematical model of the structural system, solving the equations represeting the physical behavior represented by the model, and interpreting the results within the assumptions of the model. Modern earthquake engineering research and professional practice cannot rely on only physical simulation or computational simulation alone. They are inextricably linked and closely dependent on each other for improving knowledge and practice, and ultimately the value of earthquake engineering for society. There are two broad needs for computational simulation in earthquake engineering. The first is research focused: using simulation to improve understanding of structural systems, geotechnical systems, and tsunamis. Traditionally, fundamental understanding has been obtained by experimentation because computational models have not been realistic enough, and research is necessary before computation can be relied to predict behavior and system performance. The second need for computational simulation is for engineering design: use of computational simulation to obtain estimates of performance as a basis for design decisions. Most engineering design involves analysis, but at this time advanced computational simulation to include the nonlinear behavior and realistic failure modes of structures and soil systems is generally not employed in structural design. The two aspects of simulation, research and design, should not be decoupled, because the research needs to drive the practice, but the practice needs to inform the research. Towards this goal, the presentation summarizes recent advances and simulation activities as part of of the NEES information technology program, NEESit.

Researchers from the University of Minnesota (UMN), Iowa State University and the University of Puerto Rico at Mayagüez and a consulting engineer from the Nakaki Bashaw Group, Inc. in California have undertaken a collaborative PreNEESR project since 2004. This project focuses on analytical and experimental efforts to investigate the behavior of nonrectangular structural walls subjected to multi-directional loading. The experimental component of the study utilizes the NEES Multi-Axial Subassemblage Testing (MAST) system at Minnesota (NEES@Minnesota). To date, large-scale testing of a T-shaped concrete wall and three rectangular walls has been completed. In each test, a variety of telepresence tools, including those developed and supported by the NEES Cyberinfrastructure Center were used to enable effective collaboration between the researchers at the test site at NEES@Minnesota and those at the remote locations. The author of this paper was on sabbatical leave at the University of Auckland in New Zealand during testing of two rectangular walls. Consequently, the telepresence capabilities of NZNEES@Auckland together with the NEESit tools were used to continue the research project without compromising the real-time collaboration between researchers during the tests. The wall tests conducted to date at NEES@Minnesota were the first NEES experiments to extensively use the telepresence tools and experience the benefits of real time capabilities to advance the simulation models. Consequently, these tests served as a testbed for various telepresence tools. While the benefits of real-time collaboration were realized satisfactorily in the experiments, improvements to the capabilities of telepresence tools were identified and recommended to NEESit. This paper provides a researcher's experience in remotely participating in seismic tests, including the ability to interact with both people and equipment and manipulate test data in real time, and describes how different telepresence technologies were used in the project as well as some recommendations made towards improving capabilities of telepresence tools to enhance real-time research collaboration.

The US Federal Emergency Management Agency (FEMA) is supporting the Applied Technology Council (ATC) to develop a next generation of performance-based seismic design guidelines for buildings. The project, ATC-58 will result in the capability to base design decisions explicitly on the expected performance of the building in future earthquake events. Performance is measured as the risk of experiencing three types of losses: 1 Deaths and serious injuries. 2 Direct capital losses attributable to damage contents. 3 Indirect losses attributable to the loss of use for repair or reconstruction. This characterization of performance provides a common framework for design decisions that can be readily adapted to the perspectives of all stakeholders (e.g. owner, architect, engineer, contractor, regulator, lender, insurer). It also facilitates efficient planning and design decisions through comparison of costs and benefits of various alternatives. A computerized basic loss assessment procedure has been developed comprising four steps. Shaking hazard is defined to specify how intense shaking might be and how often it is likely to occur. Corresponding ground motions are applied to a nonlinear model of the building to calculate the demands (forces and distortions) on structural and nonstructural components and systems produced by each ground motion. Up to this point the procedures are somewhat similar to those that have been used by engineers for design purposes. The next steps depart from traditional design procedures. The demand is used to generate an estimate of damage to the building including damage to the structure, to nonstructural components and systems, and to contents. Finally, the projected losses are determined based on the total damage and consequence functions that translate the damage into human losses, capital losses, and downtime. These steps have been programmed into a Performance Assessment Calculation Tool (PACT) that will be released in the Spring of 2007. This paper describes the unique types of data required for the new procedures that are related to the damage and loss calculations within PACT.

A zipper frame is a conventional braced frame with chevron type braces at all floors, connected through a zipper column linking their midpoints. Under lateral force, a 3 stories frame is expected to behave as follows: after the 1st floor braces buckle, the zipper column transmits the unbalance vertical force to the 2nd story, forcing them to buckle. The new unbalanced vertical force is then transmitted to the third story, which is designed to remain elastic. This is known as the zipper mechanism. Three shaking table tests were performed to evaluate the seismic response of a 3 story suspended zipper frame model. In every shaking table test performed, the braces buckled out of plane. It was also found that the 1st floor brace to beam gusset plate was a very important detail that affects the development of the zipper behavior. If the gusset is too stiff, the out of plane moment is transmitted to the beam, twisting it, yielding in torsion and interrupting the zipper mechanism. On the other hand, if it is flexible, the gussets bends and the beam remains straight while the zipper mechanism develops. A new analytical model was developed to capture the three-dimensional phenomena of buckling observed in the testing, while also capturing the local behavior of connections. The model is based on the well known corotational formulation of frames. Every element is described independently. Rigid body modes are removed from the analysis by using a corotated coordinate system. Within the deformed shape, the element can undergo large rotations and large strains, which are evaluated in respect to the corotated origin. The structural problem is solved in the state space, where displacements as well as velocities are variables. The final set of equations is a combination of differential and algebraic equations that is solved using the IDA software. This paper presents the main observations from the shaking table tests as well as the general description of the proposed analytical formulation and examples.

Protection against collapse is a primary objective of earthquake resistant design. In recent years much effort has been devoted to the development of reliable analytical tools that predict global collapse of structural systems due to dynamic instability. This instability is usually caused by a combination of P-Delta effects and deterioration in strength and stiffness of structural components. Lack of data to model deterioration properties of structural components does not enable us to predict collapse in a systematic and reliable manner. This paper focuses on the development of an extensive database on deterioration properties of steel beams and columns subjected to cyclic bending moments. This database is based on monotonic and cyclic component experiments for steel beams and columns performed over the last forty years. It provides relationships that associate analytical model parameters such as cap strength and deformation, post capping stiffness, and residual strength with detailing criteria that control deterioration of components. The database is used to obtain central values and measures of dispersion of all parameters needed to describe the cyclic deterioration behavior of steel beams and columns. The deterioration parameters of each component of an example steel frame structure are extracted from the steel database and a collapse prediction for the building is performed.

As the result of research conducted under a NEESR award, new design procedures are being proposed for inclusion in the 2010 AISC Seismic Provisions for Structural Steel Buildings for the design of zipper braced frames. A zipper frame is a conventional chevron-braced configuration in which the intersection points of the diagonals have been linked with a vertical element (zipper strut). The intent of the vertical elelment is to force simulataneous buckling of all stories to avoid the soft-story mechanism typical of concentrically braced frames. Zipper frames have been considered as a viable alternative to conventional braced frames in the commentary of the AISC Provisions since their first edition. However, there have been neither specific design recommendations nor definitive studies on the seismic performance of this structural configuration. Based on extensive analytical and experimental work conducted jointly by Georgia Tech, the U. at Buffalo, the U. of Colorado-Boulder and the U. of California-Berkeley, a new design procedure has been developed to ensure ductile and reliable behavior of this structural system. This paper briefly reviews that design procedure and illustrates its application through the design of a 9-story frame used in a critical facility in a high seismic area.

Recent advances in modeling fatigue and fracture in steel structures are outlined. Current approaches to simulate earthquake induced fracture and fatigue in steel structures rely on experimental and semi-empirical methods, or conventional fracture mechanics. While the semi-empirical methods cannot be generalized to a wide range of structural details, conventional fracture mechanics can be reliably used only to simulate brittle fractures similar to those observed during the Northridge earthquake, where large scale yielding is absent. The physics-based micro-models similar to the one described in this paper seek to simulate the fundamental processes of void growth and coalescence and granular cleavage responsible for fracture and ultra low cycle fatigue in structures. These models are relatively free from assumptions regarding behavior and can be used with accuracy to simulate fracture and fatigue in a general sense under a variety of conditions. Either void growth or cleavage can cause sudden crack propagation and strength deterioration at the global structural component scale. Thus, these micro-models, relying on fundamental physics are equally applicable to situations that are regarded as "brittle" or "ductile" at the structural or component level. Examples (including information from a recent Network for Earthquake Engineering Simulation and Research — NEESR project) to illustrate the use of one such model — the Cyclic Void Growth Model (CVGM) to simulate fracture through continuum finite element analyses are presented. Areas for refinement of the models and future work are outlined.

Steel moment connections in the US have been placed under very close scrutiny after the 1994 Northridge earthquake because of their unexpectedly poor performance. These investigations summarized in the FEMA-350 report produced design recommendations for new steel moment-resisting frame structures. At the core of these design recommendations is requirement that connections in Special Steel Moment Resisting Frames (SMRF) have a total rotation capacity exceeding 4% radian while maintaining strength above the nominal plastic capacity of the beam or loosing no more than 20% of the maximum connection resistance. To demonstrate this, a pre-qualification test requirement was imposed. Each new connection type is required to pass a series pre-qualification test conducted using a prescribed procedure. Such pre-qualification test approach was adopted and codified in FEMA-350 and AISC documents because rotation capacity of steel moment connections could not be reliably predicted using conventional analytical models, such as finite element or fracture mechanics models. Even though this approach is safe, it is expensive and it hampers development of new connection designs. During a severe earthquake only few cycles of large deformation may be sufficient to cause failure of steel structural components sensitive to low-cycle fatigue. Failure of pre-qualified steel moment connections after only a few cycles of high-amplitude deformation was observed in pre-qualification tests as ductile tearing of the metal in the creases of the beam buckled shape in the plastic hinge region. While low-cycle fatigue resistance is not an explicit requirement of the connection pre-qualification test procedure, resistance to this failure mode is important for seismic structural design because steel structures may have to endure a number of significant earthquakes, each having a few cycles of large deformation demand that may pose a cumulative connection fracture risk. In this paper we propose and develop a simple analytical method to predict failure of connections due to buckling and low-cycle fatigue and estimate their seismic rotation capacity in special steel moment resisting frames under both monotonic and cyclic loading conditions. The proposed method uses a plastic collapse approach and a yield-line model of a beam plastic hinge. A so-called yield-line plastic hinge (YLPH) model was developed to model local buckling of a beam in a moment connection. This approximate method was based on interpretation of experimental evidence on the shape of the buckled plastic hinge region. Both strength degradation failure induced by local buckling of beams and low-cycle fatigue failure caused by plastic strain accumulation at critical points in the plastic hinge region were considered. Then, YLPH models for common US post-Northridge connections, such as the WUF-W, the RBS and the Free Flange connection, were constructed and applied to compute their seismic rotation capacity to validate them by comparisons to experimental data.

Fracture and fatigue-induced failure in welded structural details is an important limit-state in earthquake resistant design. Despite its significance, fundamental, physics-based models to simulate Ultra Low Cycle Fatigue (ULCF) in base and weld metals are not readily available, and many of the popular approaches predict ULCF in an empirical manner without considering the complex interactions of stress and strain histories responsible for it. While convenient, these empirical methods may not be reliably transferable to untested details or connections. In this paper, newly developed physics-based models that aim to simulate ULCF at a continuum level (and apply them through finite element analyses) are introduced. Preliminary results from experiments on six column base plate specimens are presented. These tests, part of a NEESR project, seek to validate these physics-based models. From a practical standpoint, these experiments provide important insights into modes and hierarchies of failure of column base plate details, especially fracture originating in the welds and heat affected zone. The parameters considered include variations in cyclic loading histories and in weld details similar to configurations commonly used in engineering practice.

Bridges are a vital component of transportation facilities. Because it was revealed in past earthquakes that bridges were extremely vulnerable to seismic effects, a project on the failure mechanism of bridges was identified as one of the high priority areas of the National Institute of Earth Science and Disaster Prevention (NIED). Consequently, a five year research project on the large-scale experiments of bridges was launched at NIED in 2005. E-Defense is the world largest and powerful shake table constructed by NIED for clarifying the failure mechanism of structural components and systems under extreme ground motions, and it started its full service in 2006. Large scale experiments using E-Defense on bridge is scheduled in 2007–2009 based on NEES/E-Defense collaboration. Two types of model are proposed for bridge experiments; component model (C1 model) and system model (C2 model). C1 models are reinforced concrete columns which are tested to clarify failure mechanism of columns which collapsed or suffered extensive damage in 1995 Kobe, Japan earthquake, and verify the effectiveness of current seismic retrofit measures and seismic performace of columns designed based on the current design criteria. Reinforced concrete columns with diamter of nearly 2 m are excited by E-Defense as shown in Figure 1. C2 models are system models which are tested to clarfify progressive failure mechanism, effectiveness of advanced technology. Several pre E-Defense studies were conducted in 2005 and 2006 to evaluate necessary information prior to the large scale experiment using E-Defense. This paper shows an experimental clarification of failure mechanism on reinforced bridge columns which suffured extensive shear failure. This was the major damage of bridges in 1995 Kobe earthquake.

Since the 1971 San Fernando, CA, earthquake, a significant amount of research has been conducted on the ductility capacity of reinforced concrete bridge columns, resulting in significant advances in the seismic design of bridges. Since most tests have been done in static or quasistatic conditions and unidirectional loading conditions, however, no method that properly evaluates the effect of multidirectional dynamic loading has been developed, and design recommendations on this effect are still limited. It is essential to evaluate the effect of multidirectional loading on the dynamic response and seismic performance of reinforced concrete bridge columns in order to develop more advanced and reliable design procedures. Research on the dynamic behavior of reinforced concrete bridge columns under multidirectional loading is still limited due to the limitation of capacity of research facilities. Mahin and Hachem conducted a series of shake table tests of circular reinforced concrete bridge columns. The effect of multidirectional loading was investigated for 406 mm-circular columns. Nishida and Unjoh conducted a series of shake table tests for three types of cross sections, circular (600 mm in diameter), square (600 x 600 mm) and rectangular (450 x 800 mm), under a near field ground motion, and then conducted dynamic analyses to investigate the effect of bilateral loading. Sakai and Mahin conducted shake table tests of a circular reinforced concrete bridge column as a part of a research project to develop a new method that mitigates post-earthquake residual displacements. Although these tests provide valuable data for development of advanced analytical procedures and findings to be considered in seismic design on the multidirectional dynamic loading effects, most columns were tested under a near field ground motion, which has a few dominant pulses, and the dynamic behavior under repetition of strong pulses has not well been investigated. In the study presented here, the dynamic response of a circular reinforced concrete bridge column specimen under a repetitive strong shaking is investigated through shaking table tests. Accuracy of an analytical model using a fiber element is evaluated with the test results, and then the effect of multidirectional seismic excitation is analytically investigated.

Experimental testing of structural systems is essential for improving knowledge about component and system performance in earthquakes. Shaking table testing can provide important experimental data about critical issues such as the effect of component damage on system response, collapse mechanisms, residual deformation and post-earthquake capacity. For example, the new E-Defense facility is a 20 m x 15 m six degree-of-freedom shaking table at the National Research Institute for Earth Science and Disaster Prevention (NIED) in Japan will provide significantly increased capability for structural experiments. Even with this facility, most structural systems are too large to test at or near full-scale. With recent improvement of the Internet all over the world, a geographically distributed testing allows researchers to combine the capabilities of two or more sites to conduct tests on structural systems that could not be performed at any one site because of capacity limits. In the United States, the George E. Brown, Jr. Network for Earthquake Engineering Simulation (NEES) is a network of experimental sites and computing resources connected by the NEESgrid system using Grid-based middleware. Under the NEES program, the Multi-site Online Simulation Test (MOST) demonstrated the use of NEESgrid for hybrid simulation, connecting two experimental sites and one or more computational sites. It showed the potential for distributed testing, but the software architecture included a coordinator that resulted in an artificial separation between the simulation model and the time integration procedure for solving the governing equations of motion. Even with these recent advances, structural testing has typically been conducted using customized software that is dependent on the configuration of an experiment and the computational procedure for the test method. Customized software, however, is difficult to adapt to other experiments, particularly when multiple sites need to communicate in a distributed test. For structural simulation, object-oriented methodologies were introduced in the beginning in 1990s. The first object-oriented analysis application was the linear and static finite element method. In 1990, Forde et al. defined classes such as Node, Material, and Element, for linear analysis. In the past decade, many researchers have developed object-oriented software for finite element analysis. In the paper, Using object-oriented software design methodologies, a framework of cooperating software classes is developed for a variety of experimental and computational approaches, allowing mixing of computational and experimental elements with communication between the two over a network.

The effects of seismic limit state performance of bearings on a whole bridge system during strong ground motions have been greatly concerned from past earthquake damages. Although rubber bearings or seismic isolation devices have replaced steel bearings since the 1995 Hyogo-ken Nanbu earthquake, they are still in use in existing bridges. This indicates that the effects of steel bearing failure on earthquake resistance of a bridge are a crucial issue. There are several physical models for steel bearing required in developing dynamic seismic response analysis model of a bridge system. The easiest way may be in such a manner that fixed and movable bearings are modeled as a nodal in a frame structure analysis. Bi-linear models are sometimes employed to represent bearing performance. A tri-linear model for fixed and movable bearings are developed based on real failure modes during the 1995 Hyogo-ken Nanbu earthquake. However, a systematic nonlinear model for steel bearing performance is yet to be established and geometric behavior modeling should be also discussed as well as material strength in developing a nonlinear model. Under these circumstances, a seismic hybrid experiment can be a powerful toll to examine earthquake response nature of a total bridge system. Seismic hybrid tests have been extensively used in earthquake resistance study of bridges, focusing on nonlinear performance of a reinforced concrete column, a steel column and various bearing devices such as a seismic isolation bearing. However, no seismic hybrid experiments have been conducted on nonlinear behavior of steel bearing so far. The objective of the present study is then to experimentally discuss the effect of near limit state in regard to fixed steel bearing on seismic performance of a total bridge system. First, we will describe a general earthquake response of a target bridge. Second, we will explain the hybrid seismic test. Third, test results will be discussed in view of superstructure response, restoring force characteristics and others. Finally, concluding remarks throughout the present study will be summarized. The terms `fixed' and `steel' will be dropped unless confusion arises in the present paper.

Recent advances in performance-based earthquake engineering stemming from research by the Pacific Earthquake Engineering Research (PEER) Center have made critical assessment of highway bridge seismic vulnerability possible. Performance of the predominant types of reinforced concrete highway bridges is integral to the evaluation of transportation network performance under high-intensity earthquake scenarios. Additionally, probabilistic quantification of bridge response and vulnerability provides insight into the evaluation of current designs and proposed technologies at varying levels of seismic hazard and different site conditions. This paper illustrates the seismic vulnerability of two benchmark reinforced concrete bridge types, typical of new construction in California. Each bridge type has a variety of column designs for different seismic demands. A model of the structures is created that accounts for nonlinear behavior of the columns, deck, abutments, and expansion joints at the abutments. Seismic demand models are then developed using nonlinear time history analysis, including under near-field excitation. Damage in the bridge components is determined from experimental tests and component capacities. Finally, approximate repair costs are estimated from both discrete bridge-level damage states and the assembly of discrete component-level damage states using unit repair cost estimate data. The procedure for the testbed bridges is developed in a modular fashion to allow incorporation of emerging structural components, technologies, and analysis methods. The vulnerability of the base bridge types is presented in this paper as a benchmark with which to compare the use of enhanced performance structural elements, demands due to liquefaction and lateral spreading, and to observe system-level performance due to experimentally calibrated column, abutment, and foundation components from other PEER researchers. Examples are provided that illustrate the incorporation of alternate analysis methods and a fully coupled soil-structure system.

In this paper the performance of a bridge underlain by liquefiable soils is evaluated in the PEER framework for performance-based earthquake engineering. To this purpose an OpenSees model of a bridge, its foundations, and the soils that support it is developed. The model is capable of capturing the response of the site including pore pressure generation and dissipation in liquefiable soils along with the effects of those phenomena on: i) the stiffness and strength of the soils, ii) the interaction of the pile foundations with the liquefiable soils including the effects of pore pressure changes on that interaction, and iii) the response of the above-ground portions of the structure. Using motions for four hazard levels, record-to-record uncertainty is evaluated by means of "cloud" analyses and efficient intensity measures for cases with and without liquefaction are identified. A framework for expressing response in liquefiable soils as the product of the response in the absence of liquefaction and a "liquefaction multiplier" is developed. The method shows potential for reduction of dispersion by using different measures of ground motion intensity for non-liquefaction EDPs and a liquefaction multiplier.

Reinforced concrete bridges located in regions of high seismicity are supported on highly ductile columns to avoid collapse during strong ground shaking. While conventionally designed reinforced concrete bridge columns are likely to ensure life safety, large residual displacements may exist following extreme earthquakes, necessitating long-term closure of highways while expensive repairs or even complete replacement is carried out. Thus, mitigation of post-earthquake residual displacements has become a major concern. To facilitate the implementation of new designs and potentially new technologies to mitigate post-earthquake residual displacements, a thorough evaluation of their impact is needed. A team of researchers working within the Pacific Earthquake Engineering Research (PEER) Center has been conducting such an evaluation using the PEER Performance-based Earthquake Engineering (PBEE) assessment methodology. This work has involved experimental validation, numerical modeling developments and application of the PEER PBEE methodology. This paper presents an overview of some of the results of this collaborative research, including a series of earthquake simulator tests carried out to investigate the effectiveness of the proposed design approach, numerical modeling for the prediction of residual displacements, and a case study comparison of a bridge designed according to Caltrans standards having reinforced concrete columns in one case and self-centering columns in another, evaluated using the PEER PBEE methodology.

Recent developments in bridge vulnerability assessment have provided insight on likely vulnerable classes of bridges. However, there is a lack of understanding of the impact of retrofit on the vulnerability, or fragility, of these classes of bridges in order to support decision making and risk reduction efforts. This paper presents a general methodology for the development of retrofitted bridge fragility curves considering the impact of various retrofit measures (elastomeric bearings, restrainer cables, seat extenders, steel jackets, and shear keys) on the performance of the overall bridge system. The implementation and practical application of the retrofitted bridge fragility curves is highlighted. These curves offer valuable insight to decision makers and bridge owners seeking to reduce the risk of damage to bridges and consequences of bridge damage on the performance of the transportation network. Analysis of the fragilities provides guidance as to the most effective retrofit for a given bridge type and performance objective. In addition to this performance-based retrofit assessment, the fragilities are incorporated in cost-benefit studies as well as a seismic risk assessment, which exemplify their importance in supporting effective investment of funds and evaluation of the impact of retrofit on regional consequences.

Post-earthquake repair costs are useful for evaluating different bridge design options in a performance-based setting. For example, expected post-earthquake repair cost estimates together with the estimated cost of new construction provide transportation managers with a ratio useful for inventory assessment and post-disaster decision support. The derivation of post-earthquake repair cost involves assessment of expected site seismicity, bridge configuration and geometry, structure seismic response, failure modes, and available repair methods and associated costs. The repair cost probability distribution is derived using first and second moments from a simple graphical tool called Fourway. This tool uses three probabilistic models to evaluate repair costs based on the Pacific Earthquake Engineering Research Center's performance-based earthquake engineering framework: (1) a probabilistic seismic demand model relating earthquake intensity measure to bridge response parameters, (2) a probabilistic damage model relating bridge demand to discrete damage states of key bridge components, and (3) a cost model relating damage states to the cost of available repair methods. This procedure is used to compare the post-earthquake repair costs of four design options for a reinforced concrete highway overpass bridge in California with four single-column bents. Bridge design options are represented by different column heights and circular and oblong column cross-sections with different dimensions. The column choices produce different post-earthquake repair costs due to differing demand model parameters for each structural component. The total repair cost for each design option is computed by combining the individual costs for each component. Repair cost ratios are used to compare the design options.

It is widely accepted that flexible foundations affect the responses of structures beneficially or detrimentally depending on the nature of input ground motion, the effective response periods and hysteretic and radiation damping of soil-foundation domain. In most analytical procedures, the effects of flexible foundation are accounted for by using lumped springs the properties of which are obtained from empirical data and relationships. Due to the wide variation in soil material characteristics and foundation configuration, the simplified approach may not represent the real performance with acceptable accuracy. In this paper, a typical bridge representative of the Central and Eastern U.S. inventory is analyzed with consideration of soil-structure interaction where embankments, abutments, and foundations are modeled in OpenSees (PEER Center) with three-dimensional finite elements. The bridge structure is modeled in Zeus-NL (MAE Center) with fiber frame elements. The two analysis platforms are combined through the multi-site multi-platform hybrid simulation framework, UI-SIMCOR. The fragility curves of the bridge are derived based on a most realistic analytical approach.

Controlled rocking of steel braced frames has been proposed for the seismic retrofit of structures. The controlled rocking approach allows a frame to uplift from its support at the base-of-column to foundation interface while displacement-based steel yielding devices are implemented at the base location to control the response. Past studies have focused on analytical investigations of the response of such a system along with design applications for controlled rocking bridge piers. This paper discusses shake table testing of a 4-legged, steel braced pier representative of a 1/5 length scale highway bridge pier. The model is subjected to a series of seismic excitations about one of its primary orthogonal axes, using horizontal and vertical base motions, and rotated 45°. to investigate pier behavior that would be expected from bi-directional horizontal base inputs. Ground motions from the 1940 El Centro and 1994 Northridge earthquakes are used along with a synthetically generated motion. The steel yielding devices used during testing are triangular plates that yield in flexure as the pier uplifts and rocks at its base. Results of the testing focus on the overall behavior of the controlled rocking pier including global hysteretic response, uplifting displacements, and pier forces.

Significant impact of seasonally frozen soil on the dynamic behavior of foundations has been found in previous studies. Presented here is an analytical investigation on the effect of seasonally frozen soil on the seismic behavior of soil-pile-bridge bent system. Elastic-plastic Finite Element (FE) analyses of a soil-pile model have been conducted to obtain the cyclic behavior of the soil-pile system under unfrozen and frozen soil conditions. The widely-used soil spring coefficients have been derived from the analysis results. Furthermore, a simplified FE model of the soil-pile-bridge bent system was created by representing the soil-pile system with the derived soil springs. The bent model was calibrated with previous results obtained from recorded data. The modal and push-over analyses were conducted with the calibrated model to investigate the effects of seasonally frozen soil on the behavior of typical bents selected from the bridge. Significant change in the stiffness and damping properties of the soil-pile system is found when the ground surface is frozen. The frozen soil has quite different impact on the dynamic properties of different bents; the influence due to frozen soil increases with decreasing bent height-span ratio or increasing overall bent stiffness. The lateral displacement capacity decreases, or alternatively, the shear demand increases during the soil freezing. Furthermore, the effect of frozen ground on the shear demand decreases with the development of plastic deformation.

This paper investigates the dynamic response of shallow foundations on linear and nonlinear soil medium using finite element method. The study was motivated by the need to develop macroscopic foundation models that can realistically capture the nonlinear behavior and energy dissipation mechanism of shallow foundations. An infinitely long strip foundation resting on soil half-space is analyzed in depth to evaluate the dependence of its dynamic responses on various parameters, e.g. foundation width, material properties, input motion amplitude and frequency etc. Special attentions are paid to choose appropriate domain scale, mesh size and boundary conditions so as to minimize the often observed numerical oscillations when the outgoing waves are contaminated by the reflecting waves at boundaries. Such judicious choice results in an excellent agreement between the finite element analysis and the analytical solution of strip foundation on linear soil half-space. Closed-form formulas are developed to describe the frequency-dependent linear dynamic stiffness of strip foundation along both horizontal and vertical directions. Various nonlinear constitutive models of soil, which exhibit the yielding and kinematic hardening behavior of soil, are implemented in this study to evaluate the dynamic stiffness of strip foundation sitting on nonlinear soil medium. The finite element analyses reveal the strong dependency of response on input motion amplitude, frequency and yielding of soil. A nonlinearity indicator is developed to incorporate the combined effects of initial elastic stiffness, yielding stress and excitation amplitude. The numerical analyses presented here provide improved understanding on the nonlinear behavior and energy dissipation mechanism of shallow foundations under dynamic loads.

The focus of this study is the development of analytical and numerical models of soil-structure interaction in support structures that are frequently used in highway construction in urban settings. In a series of field tests, we have collected detailed and high-resolution data involving small and large-diameter reinforced concrete piles, and pile groups. The data were used to develop and validate simplified (macroelement), as well as detailed 3-dimensional finite element models of these support structures. Our findings indicate that pre-existing analysis methods and design guidelines do not accurately address the need to capture the broad variety of responses that are encountered in seismic regions due to different boundary conditions and structural geometries. Our newly developed models aim to close this perceived void, and are amenable to be routinely used in common design and analysis software used by practicing engineers.

Reinforced Concrete (RC) bridge columns could be subjected to combined flexural, axial, shear and torsional loading during earthquake excitations. This fact is particularly true for bridges that are skewed, curved, have unequal spans, or unequal column heights. In addition, multi-directional earthquake motions, significant vertical motions, and structural constraints due to stiff decking, movement of joints, abutment restraints, and soil conditions may also lead to combined loading effects. This combination of seismic loading can result in complex flexural and shear failure of these bridge columns. As part of a NEES-R funded study, a total of 14 columns are being tested under various loading conditions: cyclic bending, cyclic torsion, biaxial bending, and combined cyclic bending and torsion. This paper presents the results of the first three columns tested under pure cyclic bending, pure cyclic torsion, and combined cyclic bending and torsion respectively. The effects of combined loading on the hysteretic load-deformation response, reinforcement stress variations, and plastic hinge characteristics are discussed. Finite element models of the RC columns were developed and analyzed for different combinations of bending and torsion to guide the experimental study. The finite element predictions are compared with the experimental data and the results are discussed. Based on the test results, it is concluded that combined loading reduces the flexural and torsional capacity of a column and can also change the failure modes and deformation characteristics.

Carbon fiber reinforced polymer (CFRP) composites have been shown to be particularly well suited for external strengthening of reinforced concrete members. The superior mechanical and physical properties of fiber reinforced polymers make them excellent candidates for repair and retrofit of steel structures as well. However, the primary application for CFRP to date has mainly been flexural strengthening. This paper reports on an analytical and experimental study conducted to investigate an innovative technology to improve the buckling behavior of steel members under compression. In this study, rectangular steel bars are sandwiched between thin mortar or PVC blocks and the entire system is then wrapped with CFRP sheets. The requirements to prevent buckling of the steel members are derived from equilibrium conditions and verified with test results. Small scaled tests showed significant improvement in the inelastic axial deformation capacity prior to buckling under compression and an improved load carrying capacity after buckling.

This paper considers UK practise for the strengthening of old cast and wrought railway structures using carbon fibre reinforced polymer composite (CFRP). It examines the nature of both metals and presents case studies of strengthened cast iron structures. It explains the potential problem with wrought iron and describes recent research which has shown that subject to certain precautions CFRP can be used to strengthen wrought iron.

In recent years, the field of structural engineering has seen extensive advances in the application of fiber reinforced polymer (FRP) composite materials for repair and strengthening of structures and their components. The merits of FRP retrofit of reinforced concrete members have been well researched and documented. Relatively limited work however has been conducted investigating the use of FRP to retrofit steel members. Carbon FRP (CFRP) materials have been used to strengthen steel members, enhance fatigue or fracture performance and provide local stability for steel compression members. Harries and El-Tawil present a comprehensive summary of the state-of-the-art of steel-FRP composite systems. The present work seeks to add to this work by investigating the concept of an FRP stabilized steel member. It is proposed that through the strategic application of FRP to a steel brace member a degree of buckling restraint may be affected resulting in a Partially Buckling Restrained Brace (PBRB). The linear behavior, high strength, and stiffness of FRP composites can be applied to a steel section to increase member stability. More specifically, small amounts of FRP composites can be utilized to increase resistance to flange and/or web local buckling. The purpose of such an application is not necessarily to increase load-carrying capacity but to restrict plastic flow of the plate member. A recent analytical study [Accord et al. 2006] utilized nonlinear finite element analysis to examine the effects of low modulus glass FRP (GFRP) strips bonded to I-shaped sections developing plastic hinges under a moment-gradient loading. The inclusion of GFRP strips provided effective bracing of the flange outstands delaying the formation of local buckling of the compression flange, ultimately increasing structural ductility. Similarly, Shaat and Fam [2006] focused on the increase in axial strength and stiffness of long and short hollow square section (HSS) columns wrapped with CFRP of varying types, number of layers, and fiber orientations. Results indicated an increase in axial load capacity for the short columns ranging from 8% to 18% while the stiffness increase ranged from 21% to 28%. A reported 13% to 23% increase in axial strength of the long specimens showed no correlation with CFRP applied. The authors attribute the lack of correlation to variation of specimen out-of-straightness and minor misalignment of the test set-up.

The research community has made great progress identifying and quantifying the characteristics of fiber reinforced polymer (FRP) composite materials used in infrastructure applications. A growing number of demonstration projects and commercial applications contribute to the knowledge base in this area. The use of FRP materials in infrastructure is still in its infancy, however, and a number of issues remain to be adequately addressed. In some cases, the dearth of understanding and/or design guidance represents a significant limitation to the broader implementation of FRP materials in infrastructure applications. To partially address these issues, a Workshop was convened in October 2004 the objective of which was to identify and prioritize research areas and issues which industry, practitioners and academia identify as requiring further attention in order to improve our understanding of the behavior of FRP materials and FRP structural and repair systems. The Workshop focused on the application of FRP materials in concrete and masonry structures. Since this time, significant interest has developed in the application of FRP for the repair and strengthening of steel structures. This paper reports the primary findings of the Workshop on Research in FRP Composites in Concrete Construction and extends these to some degree into the emerging area of FRP Composites in Steel Construction. In this work, the discussion of research needs should not be construed as representing an impediment to the broader adoption of FRP materials for infrastructure applications. Rather, this discussion of research needs acknowledges that, as with any emerging technology, limitations and gaps in knowledge exist. Considerable effort has been made to identify these gaps and, in some sense, to prioritize their importance with respect to the broader adoption of FRP materials. A clear, unified understanding of research needs is necessary for this emerging technology to become mature.

Coupled core walls (CCWs) are attractive lateral-force resisting systems with remarkable lateral stiffness. The extent of the lateral stiffness depends on the type of coupling beams used to couple the wall piers. Current building codes essentially imply the use of diagonally-reinforced coupling beams with practical span-to-depth ratio of two to four. Inherent difficulties associated with design and construction of diagonally-reinforced coupling beams are major impediments for CCWs, and are often cited as one of the primary reasons for selecting other systems despite the many advantages that CCWs offer. Performance-based design (PBD) paradigms and/or innovative systems offer viable alternatives to overcome these difficulties. This paper provides an overview of these two options.

The ASCE Committee on Composite Construction (CCC) has been working for two years to develop recommended seismic design guidelines for hybrid coupled walls, which are comprised of two or more reinforced concrete walls connected with steel beams distributed along the height of the walls. The effort is being conducted by a task group comprised of 7 individuals funded by an ASCE Special Project. The recommendations developed as a result of this effort can be used for proportioning HCW systems in zones of moderate to severe seismic risk. This paper summarizes the key features of the document produced by the CCC task group, which is currently nearing completion. Readers should note that at the time this paper was written, the developed document had not yet been reviewed or approved by ASCE and therefore the contents described herein may differ from the final document.

Performance of buildings under seismic loading is often decided based on damage states observed in the structural and non-structural members. The type and extent of damage states exhibit a wide range depending on the structural system resisting the earthquake loading. While developing performance-based design methodologies, it is critical to provide computational tools to assess the available capacity of structural members corresponding to various performance-levels (e.g., Immediate Occupancy, Life Safety). In this paper the progression of local damage states (e.g., concrete crushing, steel yielding) of rectangular concrete-filled steel tube (RCFT) members and the effect of local damage states on the structural performance of composite RCFT frames was investigated using a new distributed plasticity finite element formulation. The computational findings of the detailed structural performance, particularly the interaction of the local damage states, were compared with experiments, as well as to damage indices of RCFT members and frames proposed by. Adopting the local damage states as the main design parameters, a performance-based design methodology is introduced and illustrated for the RCFT beam-columns of a three-story composite RCFT frame.

The behavior and seismic performance of composite moment resisting frames (MRFs) comprised of concrete filled tubes (CFTs) and wide flange beams was experimentally investigated. A four-story two-thirds scale composite MRF test structure was designed using performance-based design concepts. The performance objectives included operational conditions after subjecting the structure to a series of ground motions representing a frequent occurrence event (FOE), life safety after a design basis event (DBE), and collapse prevention when subjected to the maximum considered event (MCE). The hybrid pseudo-dynamic test method was used to subject the test structure to these various seismic hazard levels. P-Δ effects associated with the gravity frames in the prototype building were analytically considered during the tests. A displacement convergence loop was included in the testing algorithm in order to achieve stable response during the tests, where inelastic behavior and participation of higher modes occurred. Results from the tests indicated that the structural performance under the simulated seismic loading was consistent with the expected performance for all the hazard levels, indicating that effective seismic performance of composite MRFs with CFT columns can be achieved.

This paper presents an experimental approach for determining the fundamental moment-curvature-temperature (M-φ-T) behavior of CFT members subjected to combined axial, thermal, and flexural loading. The experimental approach involves the use of: (a) innovative heating and control equipment for applying the thermal loading, and (b) digital imaging system for measuring deformations at elevated temperatures, both of which are presented in detail. The experimental results for three CFT beam-column specimens are presented in detail. These specimens were subjected to different axial load levels (15 % and 30% of the axial load capacity) and two hours of heating subjecting the steel surface to temperatures of about 500°C. The experimental results illustrate the effects of heating and different axial load levels on the fundamental behavior of the CFT members.

Most Fibre Reinforced Polymer (FRP) systems for strengthening concrete structures use epoxy resins. Chemically epoxy resins are plastics. In comparison to concrete epoxy resins show material properties more sensitive to time and environment, for example loss of strength or creeping under sustained loading. Regarding applications under elevated temperatures or strengthening systems creating significant permanent bond stresses in the adhesive — almost all modern, pre-stressed uses of FRP materials act like this — the performance of the system can be limited by the material properties of the epoxy resin. The interaction of the environmental conditions and the stress situation in the adhesive layer with the system has to be known for a safe, economical and durable design. This paper reports about experimental long — term tests on the properties of epoxy resin bonded Near Surface Mounted (NSM) Fibre Reinforced Polymer systems. Basing on these experimental results a calculation model is derived and applied to examine the development of crack widths, transfer lengths and tension stiffening effects in strengthened concrete structures.

In this research project four post-tensioned slab specimens were tested. Two of the specimens were simply supported, two-way slabs with two-way post-tensioning. One of the specimens was a one-way slab with fixed ends and one-way post-tensioning. The last specimen was a one-way slab with one fixed end and one simply supported end and one way post-tensioning. Part of this research project was focused on the study of cracking, deflection and ultimate strength behavior of un-bonded post-tensioned slab specimens. Loading was stopped prior to collapse and the heavily cracked slabs were saved. The main objective of this investigation was to repair these slabs with CFRP (Carbon Fiber Reinforced Polymer) composite materials and find the flexural capacity, cracking pattern and general behavior. Two simple supported two-way slabs with similar un-bonded post-tensioning arrangement and with different mild steel were tested with uniform load. The approximate dimension of the slab was 9' ½" x 9' ½" x 3" depth. Sixteen ¼" diameter un-bonded post-tensioning strands were placed uniformly in each direction. The post tensioned strands had a parabolic drape and were stressed to appropriately 70% of ultimate strength of the strands. Theoretical estimate was made for cracking at elastic stage and at collapse stage using Yield Line Theory. The cracked slabs were externally repaired and strengthened by using CFRP fabric and adhesives. Two different strengthening schemes were used and compared. The repaired slabs were tested again to collect similar data as it was done for the original slabs. Comparisons were made between the repaired and control specimens and the measured values and estimated values of the tested slabs. The support conditions for the two one-way slabs were varied. Sixteen un-bonded post-tensioning strands and mild steel reinforcement were placed in one direction only. Other details and testing procedure remained similar to that of the two-way slabs.

For applications in civil infrastructure rehabilitation and new construction, wood is being reinforced with externally bonded laminates and fabrics of fiber-reinforced polymer (FRP) composites. The potential benefits of external reinforcement with bonded FRP composites have been demonstrated through several studies, and the field implementation of this technology has been successfully proven in numerous projects. However, there is a concern with the reliable performance of the interface bond, since an inadequate bond strength and integrity can render the reinforcement ineffective and lead to premature failure of the structure. The service performance of wood-FRP bonded interfaces under static, environmental and cyclic loads is presented. The durability and shear strength of wood-FRP is evaluated by modified ASTM tests. An innovative contoured double cantilever beam (CDCB) and tapered end-notched flexure (TENF) specimen is used to evaluate the mode-I and mode II fracture and fatigue of wood-FRP bonded interfaces. Based on this study, recommendations and guidelines are given for characterization and evaluation of wood-FRP bonded interfaces. The proposed approach can further be extended to characterization of other bonded interfaces for dissimilar materials.

Current approaches to building design for large lateral seismic demands typically involve the use of ductile structural systems, which in most cases utilize the gravity load-carrying members to resist the lateral movement. The resulting inelastic behavior reduces the design base shear while preventing collapse and allows for economic design of the structural components. However, inelastic behavior leads to structural damage, which in most ductile frame lateral systems means damage to the gravity load-carrying members. The loss of occupancy and the difficulty associated with repairing the gravity system economically burdens the owners and occupants. A structural steel framing system free of diagonal bracing and intended for rapid return to occupancy is outlined. The lateral system consists of dual columns interconnected with replaceable link beams and a secondary moment frame. The linked columns provide inelastic behavior to the system by having the links yield primarily in shear under predetermined levels of lateral load, while protecting the columns and beams that carry the gravity loads. Non-linear pushover analyses were used to investigate the performance of the proposed lateral system. Under increasing drift, a ductile system response was obtained with inelasticity initiated in the links. The structural system exhibited three levels of performance: 1) elastic, 2) rapid return to occupancy and 3) collapse prevention. The rapid return to occupancy can be achieved prior to yielding of the gravity beams by replacing the damaged links, offering an improved level of performance over conventional moment frames. The linked column contribution toward system stiffness and strength was found to depend on the link strength and the secondary moment frame connectivity. The system behavior was ductile with drift levels exceeding special moment resisting frames, but with reduced demand on the beam moment connections. The closely spaced linked columns can develop large axial forces that need to be addressed at the foundation level, but also resist significantly lower moments than the special moment resisting frames. In general, the non-linear analyses showed the linked column frame system to be viable under seismic loading, offering the designers and owners a ductile structural system with specific target performance levels.

In this study, the seismic response of a pair of linear multistory building structures linked to one another at floor levels by nonlinear passive devices is comprehensively evaluated. Three possible height-wise distributions of the connectors were analyzed for wide ranges of values of their characteristic constants. Two response quantities were considered: interstory drift and absolute floor acceleration. Results indicate that, while it is possible to reduce both response quantities in both buildings, the level of response reduction generally decreases with decreasing values of damper exponent α. Results also indicate that the response of the systems is not sensitive to the height-wise distribution of the connecting dampers. Finally, it was also found that the optimal value of the damper exponent α depends on the value of the sum of the damper coefficients of the linking devices (c₁+c₂= C): while the optimal value of α is 0.80 when the value of C is relatively low, the optimal value of α is 1.00 (i.e., a linear viscous damper) when the value of C is relatively high.

Composite construction incorporating steel beams and precast hollowcore slabs is a recently developed composite floor system for building construction. As the construction industry demands for rapid construction with reduction in cost and environmental impacts, this form of composite construction, which does not require major onsite concreting, has become very popular among the designers and engineers in the UK. This form of composite construction is so far limited to simple beam-column connections. A semi-rigid composite joint is developed which can provide sufficient moment and rotation capacity required for plastic analysis in composite beams design. An analytical model for the semi-rigid composite connection is proposed and is verified with the experimental data and good agreement is obtained.

Although many of the key physical phenomena associated with progressive collapse can be adequately modeled by high fidelity finite element analysis, such models are prohibitively expensive for routine design applications. The alternative is to use macro models that capture the essence of the sought structural response in a computationally efficient manner. This paper describes how suitable macro-models are developed to permit accurate modeling of progressive collapse analysis of a steel moment resisting building frame. The developed models are calibrated using detailed finite element analysis of beam-column sub-assemblages. Progressive collapse simulations of a ten-story moment resisting frame building are presented to demonstrate application of the models.

This paper discusses the sensitivity of collapse potential of buildings to variation of structural parameters in different structural systems. In this study, collapse capacity of a building subjected to a single ground motion is defined as the scalar ground motion intensity measure at which the building will become dynamically unstable. The process for obtaining the collapse capacity using Incremental Dynamic Analysis is illustrated. For a given building, collapse capacities of a number of ground motions are used to estimate the collapse fragility curve for the building. Two types of structural systems are considered in this study, moment-resisting frames, and shear walls. Simple mathematical models denoted as "generic structures" are devised to model moment-resisting frames and shear walls. A comprehensive database of collapse fragility curves (assuming a lognormal distribution for building collapse capacity) for a wide-range of combination in structural parameters of generic moment-resisting frames and generic shear walls are developed. Using this database, closed-form equations for median and dispersion of building collapse capacity curves are developed. Such equations not only facilitate the design and assessment processes, but also help in understanding the major trends and importance of certain parameters in changing the collapse potential of a structural system. Comparison between estimates of median collapse capacity using the close-form equations and data show that the presented equations are in good agreement with the data. It is concluded that the primary causes of collapse are severe cyclic deterioration and small plastic hinge rotation capacity of structural components. Furthermore, it is shown that P-Delta effect is the major cause of collapse especially in moment-resisting frames with a small yield base shear coefficient. Another parameter that greatly affects the collapse potential of moment-resisting frames is the ratio of column to beam strength. It is shown that increasing this parameter from 1.2 (ACI suggestion) to 2.4 could increase the median of collapse capacity by up to 90%. In this study, effect of incorporating a vector-valued ground motion intensity measure for estimating the collapse capacity is investigated. It is shown that using the scalar ground motion intensity measure for defining the collapse capacity can lead to underestimation of median collapse capacity by 50%, compared to using a vector-valued intensity measure for this purpose. At the end, a brief discussion of the methods for design for collapse safety is presented.

After the 1994 Northridge earthquake that highlighted serious design and construction deficiencies in steel connections, new design guidelines were developed to remedy their performance under seismic effects. Utilizing available steel connections, unreinforced and reinforced with respect to earthquake load, this numerical study was performed to develop a better understanding as to the behavior of moment resistant steel connections under blast loads, identify possible performance deficiencies, and provide recommendations for research that could help to remedy such difficulties.

Blast effects testing and analytic modeling of the components of steel frame systems have provided data revealing that steel frame systems, while likely to behave ductility in general, have localized nonductile response mechanisms that actually dictate whether the system as a whole will perform well in a blast environment. Evidence from recent blast effects tests indicates the sections themselves perform well for car bomb type loadings under both weak and strong axis loads. In addition, analytic models indicate close agreement with test results. However, a troubling aspect of these studies is defining criteria (e.g., fracture strain) that can be correlated with the occurrence of the fractures in sections or connections. These criteria are to be used in analytic models to predict blast effect responses. The ability to predict section and connection fractures/failures and their dependence on strain rate, stress state, or other material and response related factors will ultimately dictate the adequacy of the high-fidelity physics-based (HFPB) models used to simulate blast effects behaviors for steel components and framing systems. This paper explores analytically various implications of the results of blast effects tests as they pertain to the resilience of steel frame systems. The paper will discuss the development of simplified models for predicting the response of steel beam-column connections to blast loads, using LS-DYNA continuum models.

Several studies reported that damage due to blast load can be significantly reduced by using seismic detailing because connection design and detailing affect the response of reinforced concrete members in moment resisting frames. Properly detailed connections allow concrete members to develop full flexural capacity resulting in ductile failure in flexural mode instead of brittle shear failure at the connections. The design of such connections emphasizes adequate reinforcement to confine the concrete in the joint and thereby increase the ductility and the shear force resistance of the joint. The ACI 318 Building Code has extensive requirements for the design of joints which are part of special moment resisting frames designed to resist high seismic loads. This paper evaluates the behavior of joints designed for seismic forces in special moment resisting frames.

Attacks on buildings in the United States over the past decade by terrorists have heightened concerns over the availability of proven technologies to economically design and construct multi-story buildings capable of mitigating progressive collapse when subjected to terrorist bomb blast. To resolve these concerns, and in recognition of the potential favorable relative economics of steel frame construction over reinforced concrete, the General Services Administration (GSA), Office of the Chief Architect, elected in 2004 to fund and conduct a first-ever, high-level blast and progressive collapse steel frame test program (the "GSA Test Program") to investigate the performance of conventional welded steel frame construction. Prior to 2004, GSA's progressive collapse guidelines for the design of steel frame buildings were based on analytical research only. The primary objectives of the GSA Test Program are five-fold: 1. To either validate or dispel, through actual testing and corroborative state-of-the-art analysis, the notion that reinforced concrete is preeminent over structural steel in mitigating the effects of bomb blast attack on conventional public structures. 2. To verify and validate analysis tools needed to conduct reliable blast and progressive collapse analyses in order to minimize the need for future costly and time-consuming full-scale testing. 3. To investigate the relative performance of contrasting steel frame beam-to-column connection types when subjected to the high strain rate effects of blast attack and subsequent impending progressive collapse conditions. 4. To identify steel frame beam-to-column connection systems that are reliable solutions for the mitigation of the effects of bomb blast, and the positive arrest of impending progressive collapse, for use by the GSA and other U.S. Federal government agencies. 5. To gain the knowledge necessary to modify, as necessary, a variety of steel frame connections types being used in the construction industry today, in order to be able to reliably and economically provide GSA's prescribed security expectations.

Bridge Scour is the major cause of bridge collapses and some of which have resulted in loss of life in different parts of the world. Over the past half century there have been many different types of scour detection methods and systems, however these techniques are extremely heavy requiring special equipment to operate and to transport from bridge to bridge making it expensive to use for a large number of bridges. In this paper, a low-cost and continuous remote scour monitoring system using polymer fiber optic sensors (PFOSs) and MicroElectroMechanical Systems (MEMS) hardware is presented and the sensitivity and sensing capability of the PFOS system is discussed.

Two back-propagation neural networks were applied to identify the stiffness coefficients of a highway bridge directly from its seismic responses without any eigenvalue analysis. An emulator neural network was trained to accurately predict the responses of a model structure that represents an as-built state of the highway bridge. A parameter evaluator neural network was trained to relate the response prediction error by the trained emulator neural network to the change in stiffness coefficients of the model structure, which represents various damage states of the highway bridge. An attempt is made to investigate the effect of various training data sets, or earthquake-induced structural responses, on the performance of a well trained emulator neural network. Furthermore, a new evaluation criterion, named weighted root-mean-square (RMS), is proposed for a more consistent performance measurement of the emulator neural networks that are trained under various earthquake excitations. Extensive parametric studies indicated that the two neural networks are very effective in the stiffness identification of highway bridges. An emulator neural network can be well trained with various seismic responses to give consistent performance measured by the proposed weighted RMS.

The focus of this study is the application of novel as well as existing methods of modal system identification and finite element model updating techniques to experimental data. The data are collected during a small earthquake and from ambient vibrations. The specimen structure is named the "Louis and Doris Factor Building," and is located at the UCLA campus. It was instrumented by the U.S. Geological Survey with an embedded 72-channel accelerometer network following the 1994 Northridge earthquake. This network provides continuous vibration data. The collected data were used first to estimate the modal properties of the building, which were subsequently used to update a finite element model of the structure using an improved method. The results indicate that the combined use of modal system identification and model updating techniques yields a practically useful picture of the structural state; and furthermore, this approach may be used for obtaining a calibrated mathematical model of the dynamic response of non-structural components.

The focus of this study is on the use of forced vibration testing and of finite element model updating techniques to quantify existing damage in a four-story reinforced concrete structure. The specimen structure is the so-called "Four Seasons Building," located in Sherman Oaks, California. It was damaged from shaking during the 1994 Northridge Earthquake, and is currently slated for demolition. Using the nees@UCLA mobile testing equipment, we have recorded floor accelerations as well as strains at select locations during forced vibration tests with linear and eccentric mass shakers. The data were used to update a finite element model of the structure. The updated model appears to correlate well with the observed distribution of damage throughout the building, and enables to quantify its severity.

This paper evaluates the current Nonlinear Static Procedures (NSPs) specified in the FEMA-356, ATC-40, and FEMA-440 documents using motions of a reinforced-concrete building that was strongly shaken during the 1994 Northridge earthquake. For this purpose seismic demands — peak roof displacements, floor displacements, and inter-story drifts — estimated from the NSPs are compared with the values "derived" from recorded motions. It is shown that the Capacity Spectrum Methods in ATC-40 and FEMA-440 documents may significantly underestimate the seismic demands. The Coefficient Methods in FEMA-356 and FEMA-440 documents, on the other hand, lead to much better estimates of seismic demands compared to the values "derived" from recorded motions.

Evaluation of existing reinforced concrete buildings has received considerable attention after consequences of their poor performance have been highlighted during recent earthquakes. Guidelines to perform evaluation of buildings were first published by FEMA and are becoming commonplace in structural engineering offices. The objective of this paper is to critically evaluate parameters contained in FEMA 356 that are used to construct backbone curves of existing conventionally reinforced coupling beams. Particular emphasis is placed on techniques used to estimate beam shear strength as well as deformation parameters that affect construction of these backbone curves. Modifications to existing methods are proposed to better reflect beam performance as measured through experimental testing.

This study investigates the accuracy of FEMA 356 shear and flexure modeling procedures for reinforced concrete (RC) columns and beam-column joints with poorly detailed or insufficient reinforcement. Following the FEMA 356 guidelines, generalized flexure and shear force-deformation relations were developed and compared with the experimental data from 26 column specimens and 17 beam-column joint specimens. Specifically, the measured and predicted responses were compared and evaluated: at yield displacement and the corresponding lateral load, lateral load and displacement at ultimate, and at axial load failure. In general, while the FEMA 356 models predict the lateral strength of columns reasonably well, they underestimate the shear strength of beam column joints. The predicted initial stiffness and deformations at both yield and ultimate are conservative for columns.