The research objective of this Faculty Early Career Development (CAREER) Program project is to enhance the hybrid simulation methodology to evaluate the performance of large scale structures loaded up to collapse. Hybrid simulation, which combines numerical simulation with experimental simulation of structural elements, provides a more realistic, reliable, and economical approach to testing structural systems under earthquake loads. Fundamental contributions, addressing issues related to numerical, experimental and boundary condition errors and assumptions, will be made in the development of algorithms used in testing and validation procedures. In particular, the computational platform will provide capabilities for robust implicit integration algorithms and fault-tolerant distributed control strategies to enable testing of complex structural systems utilizing geographically distributed testing of its substructures. The resulting testing and simulation platform will be used to evaluate the seismic performance of large-scale steel moment frame structures from the onset of damage through collapse. Only key subassemblies will be experimentally tested and the global system response will be captured through interaction with the numerical model. The efficacy of hybrid simulations to predict collapse will be validated by comparing the results recently obtained from full scale testing in Japan.
Upon its completion, the improved hybrid simulation platform developed in this project can be used to generate complete fragilities of structural components and systems from the onset of damage through collapse as required for next-generation performance-based earthquake engineering. It can also be used to assess the collapse safety margin of buildings inherent in current building codes. In particular, our understanding of the seismic performance of steel buildings will be improved through the large-scale hybrid testing program and numerical simulation studies. An integrated education and outreach plan will lead to development of new curriculum material and training of graduate and undergraduate students in various stages of structural design process such as preliminary design, computer simulation, design revision, construction and physical testing. The outreach activities are planned in partnership with the University at Buffalo Graduate School of Education and local schools districts, and they will include underrepresented students.
Large-scale experiments of structural systems under extreme loading are necessary to develop a better understanding of structural behavior and to advance the state-of-practice of structural design. For example, the unanticipated failures of steel connections during the 1994 Northridge earthquake exposed deficiencies in knowledge of the behavior of structures under seismic loading. Instead of waiting for the next large earthquake to find other deficiencies in the design and construction of modern structures, the behavior of structural systems can be evaluated through large-scale experiments. In fact, NSF has invested significant resources into the George E. Brown Network for Earthquake Engineering Simulation (NEES) to transform experimental earthquake-engineering research. Testing facilities include large shaking tables to test models of structural systems usually at reduced scale or reaction wall facilities that can apply large loads on structural components or subassemblies. The project has focused on the development and implementation of a hybrid approach to testing that can provide a safer, less expensive way to examine how buildings are damaged and eventually collapse during earthquakes. Hybrid simulation combines numerical simulation of substructures with predictable behavior, and experimental testing of complex components that are difficult to model. This approach has significant potential for realistic and economical experimental evaluation of complex structural systems by providing system level simulations with structural testing of only the critical components at large scale using reaction wall facilities . First to evaluate the procedure, hybrid simulation of a four-story steel moment frame building, previously tests on the largest earthquake simulator in the world (E-Defense, Japan) was carried through an internationally distributed hybrid simulation. In collaboration with international researchers, the structural model was divided into two experimental substructures, one in the University at Buffalo and the other in Kyoto University, Japan taking advantage of space and capabilities of both laboratories. The simultaneous simulation of both experimental substructures was linked to the numerical model of the remaining structure through the internet. The results of the experiments compared well to previous shake table simulations, demonstrating that hybrid simulations can be effectively used to examine the seismic response of structural systems through collapse. After validation of the testing method, hybrid simulations were applied to evaluate the performance of a common type of structural system used to resist earthquake forces. A special steel moment resisting frame and gravity framing designed to recent code standards in the US was and tested. The hybrid simulations included large scale frame subassemblies with concrete floor slab subjected to realistic loading conditions to capture the system-level response of this structural system. Data from the seismic response of the two half-scale subassemblies of a steel moment and gravity frame through collapse allowed for the observation of component behavior (girder, column, panel zones, etc.), their connections and interactions with neighboring members which resulted in improved system-level simulations. The large scale experiments also generated data for the validation of numerical simulation capabilities that will lead to improved design standards and more reliable simulation capabilities for design engineers. This same approach can be applied to other structural systems, including new systems, to experimentally evaluate their performance under earthquake loading. The main outcome of this research is the development of simulation methods and their application to understand the progress of failure in structures during earthquakes. Throughout this project, the research team has been very active in exposing middle and high school students to engineering, particularly students from underserved communities. Hands-on activities for K-12 students have been developed and organized visits by students to the university and well as visits to classrooms have involved demonstrations of shake tables and hybrid simulation. Activities have focused on having students design and construct wood models then test them to collapse using educational shake tables.
