The goal of this project is to develop validation data and nonlinear models for nonlinear response, component failure mechanisms, and internal force redistribution as collapse occurs in a building frame representative of older concrete construction.
This project supports the PEER strategic plan by providing performance data, validation tests, and nonlinear models to advance the simulation capabilities of OpenSees. Performance data and simulation are central to the PEER mission of developing performance-based earthquake engineering methodologies, in this case with particular application to the Van Nuys building testbed.
We previously have completed pseudo-static and dynamic shaking table tests of specimens sustaining shear and axial-load failures. These serve as the basis for development of analytical models as well as computational procedure.
Models for shear strength have been developed and checked against available data for representative columns tested both within PEER and by other organizations. Some comparison with test results is in Figure 1.
Models for axial-load failure following shear failure also have been developed and compared with available data. These models show a correlation between axial load, transverse reinforcement, and drift at failure. We have implemented the model in OpenSees and have validated it against results of earthquake simulation tests conducted previously within this project. Figure 2 shows comparison between measured and calculated axial load response for a shear-critical column in a three-column frame, indicating good correlation. Additional details can be found at http://peer.berkeley.edu/%7Eelwood/research/dissertation.htm.
An important finding of Year 5 research was that simulation of axial failure in building frames requires accurate simulations for lateral load response of systems with degrading strength. We have redirected some of the effort for Year 5 and Year 6 to develop additional earthquake simulator data on dynamic response of these types of systems. One set of tests planned for summer 2003 involves one-story specimens with various degrees of strength degradation. Another set of tests for later in the year extends the study to multi-story systems.
Studies of three-dimensional collapse response remains of interest in this project, but is premature at this time.
Figure 1. Ratios of measured to calculated shear
strengths
Larger View
Figure
2. Calculated and measured axial loads for three-column
frame during shear and axial failure
Larger
View
Researchers in Japan and Taiwan have carried out similar work. The details and configurations used in those tests are different from those considered common and most critical for existing US buildings. However, we remain in close contact with our counterparts in other countries so that we can take advantage of their conceptual thinking, data, and analysis models. Within PEER, we are collaborating with developers in OpenSees to ensure optimal development of analytical models and simulation modules.
We anticipate completing earthquake simulator tests during Year 6 and into Year 7. We propose to continue work in this general direction with emphasis on the development of nonlinear models/simulation strategies for systems with strength degradation. As this problem becomes further resolved, it will be important to extend the work by examining three-dimensional collapse scenarios for conventional building framing systems.
The work on shear strength of reinforced concrete columns has been incorporated in FEMA 356, which is the pre-standard for seismic rehabilitation of buildings and serves as the primary source of information for seismic evaluation and retrofit of existing buildings in the U.S. The work on seismic shear strength and on loss of axial load capacity has been used by several engineering firms for seismic assessment of existing buildings. One specific instance is work by Forell-Elsesser Engineers in their seismic evaluations of buildings on the UC Berkeley campus.
- Publication of final PEER reports on shear and axial failure (June 2003).
- Complete shaking table tests on single-story specimens (December 2003).
- Complete shaking table tests on multi-story specimens (June 2004).
- Analytical models for shear and axial load behavior
- Validation test data from single-story and multi-story shake-table tests.