Project Title/ID Number Full-Scale Frame Test Validation of Simulation and Performance Demands—4202002
Start/End Dates 10/1/02—9/30/03
Project Leader Greg Deierlein (Stanford/Faculty)
Team Members Paul Cordova (Stanford/Grad Student)
Project goals and objectives

This is a collaborative project between researchers at PEER and the National Center for Earthquake Engineering Research (NCREE) in Taiwan to utilize data from a full-scale frame test to validate structural simulation and performance models. Specific goals of the project are to

  1. assess the accuracy of OpenSees simulation models through detailed comparisons with data from a full-scale moment frame test and complementary subassembly tests,
  2. collect and evaluate test data to develop relationships between engineering demand parameters and structural damage measures, and
  3. benchmark the seismic performance of a modern (IBC 2000 compliant) three-story building frame system.
Role of this project in supporting PEER’s vision

Successful implementation of PEER’s vision for performance-based engineering will depend in large part on the extent to which PEER demonstrates the accuracy of simulation and performance models to evaluate IM-EDP-DM relationships. The full-scale component and frame test results available to this project provide an extensive and well-documented data set that is ideally suited for validating such models. As such, detailed OpenSees simulations and validation comparisons to this full-scale test data are central to PEER’s mission.

Methodology employed

Larger View Larger View

Figure 1. Comparison of test data and OpenSees simulation of pre-cast reinforced concrete column subassembly test

This project utilizes data from a full-scale (three-story three-bay) frame test, which was conducted at NCREE in October 2002. Planning, design and execution of the test were conducted with external (non-PEER) funding as a collaborative effort between the PI (Deierlein), co-PI (K.C. Tsai of NCREE), a Stanford PhD student (P. Cordova), and other NCREE researchers. In addition to the frame test itself, a number of companion subassembly tests of RC columns and RCS beam-column joints were conducted. This Year 6 PEER project provides support for the PI and his graduate student to simulate the frame test using OpenSees and utilize the test data for validation of simulation and performance models. This test data includes design drawings and calculations, material tests, full instrumentation (loads, displacements, strains), and digital photographs. This data has been collected and archived in an electronic database, which is available on-line to the research team.

The frame test is a full-scale hybrid structure, consisting of reinforced-concrete columns and steel-concrete composite beams. Measuring 12 meters tall and 21 meters long, the frame is one of the largest structures of its type ever tested. The frame was tested pseudo-dynamically under four sequential earthquake records, representing hazard return periods of 100 years, 500 years, and 2500 years. Following the pseudo-dynamic tests, the frame was subjected to a quasi-static push out to interstory drift levels of 10%. Several OpenSees models of the frame have been developed to evaluate alternative component models and parameters for columns, beams, and joints. In conjunction with detailed comparisons of the analysis and test data, efforts are underway to relate the calculated EDPs (drifts, hinge rotations, etc.) to the recorded damage (photographs, hysteretic response quantities, system identification quantities, etc.). Where appropriate, statistical methods are applied to rigorously quantify the results in a format consistent with the needs of the PEER PBEE framework.


Figure 2. Three-story three-bay full-scale frame test (courtesy, K.C. Tsai, NCREE)
Larger View

Brief description of past year’s accomplishments and more detail on expected Year 6 accomplishments

Prior to the initiation of this project in Year 6, the PI and his student collaborated with NCREE to develop the frame design (in accordance with US design standards) and participate in the frame test. Since completion of the frame test in Oct. 2002, efforts are underway to develop and validate the OpenSees model. Initial results (as of April 2003) are fairly encouraging with respect to the accuracy of OpenSees to calculate maximum (extreme) values of drift and rotation. Work is continuing to complete this project in Fall 2003.

Other similar work being conducted within and outside PEER and how this project differs

The general topic of nonlinear simulation is one of general interest and activity in the earthquake engineering research community. The specific scope of this project to validate the analysis with a full-scale frame test is unique to this project. This work is being conducted as a collaborative effort with researchers at NCREE in Taiwan. This collaboration leverages significant external support from NCREE for building and testing the frame.

Plans for Year 7 if this project is expected to be continued

This project is a one-year effort that will be completed early in Year 7. The model validation achieved during the project will help lay the groundwork for a RC building benchmarking study, which will be initiated by the PI as a separate effort in Year 7 under Thrust Area 3.

Describe any instances where you are aware that your results have been used in industry

As this is a single-year effort that just began in Year 6, results of the project are not to the point of being used by industry. It is anticipated that the successful validation study will facilitate the adoption of advanced nonlinear simulation models in structural design practice.


Figure 3. Comparison of time-history traces (test versus OpenSees simulation) of roof drift for four sequential input ground motions
Larger View

Expected milestones
  • October 2002 - preliminary analysis and pseudo-dynamic test of full-scale three-story moment frame (completed)
  • May 2003 – detailed critique of data from pre-test and post-test analysis models and frame test (in progress, March 2003)
  • October 2004 – report documenting and comparing analysis and test results, including details on OpenSees modeling attributes and seismic design implications (in progress, March 2003)

Report (and papers) documenting experimental and analysis results of full-scale frame test, including detailed comparisons to validate the OpenSees analysis models. Apart from the written report, the project will likely result in an on-line (web accessible) database of the experimental and analysis results.