Project Title/ID Number Validation of Simulation and Performance Models of RC Joints —5352002
Start/End Dates 10/1/02—9/30/03
Project Leader Dawn Lehman (UW/Faculty)
Team Members John Stanton (UW/Faculty), George Gimas (UW/Undergrad Student), Meredith Anderson (UW/Grad Student)
Project goals and objectives

The goals of the study are to develop practical simulation and performance models for beam-column joints that lack transverse reinforcement. The specific objectives, which constitute the research tasks and deliverables, include:

  1. Development of simple simulation tools for immediate use by the profession. (Task 1)
  2. Validation of more advanced simulation tools. (Task 2)
  3. Development of performance models appropriate for each type of simulation model, including acceptance criteria. (Task 3)
  4. Development of guidelines for the use of each pair of simulation and performance models. (Task 4)
Role of this project in supporting PEER’s vision

A primary focus of the PEER research effort is the development of advanced simulation and performance models for non-ductile building components. The PEER research efforts have largely focused on Multi-Component Macro-Element models, which will offer a family of tools for the research community and perhaps for the profession in the future. However, current structural analysis software and the present business environment do not permit engineers to use these tools today. Therefore, in addition to these advanced models, the research community must provide practical tools with which the profession may implement PBEE both today and in the future. The research proposed herein will develop practical tools for performance-based earthquake evaluation of older reinforced concrete structures. The research program will focus on simulation and performance tools for beam-column joints. Because retrofitting joints is very expensive, accurate and reliable evaluation of the need for retrofit is important, and proper simulation of joints is, in turn, critical. This research effort will complement the current PEER research effort by Lowes and Theiss and previous PEER research effort by Lehman, Lowes, and Stanton.

Methodology employed

Robust engineering tools are needed for performance-based earthquake engineering and therefore must be validated using analytical models (which have been validated previously), experimental results, or both. Here, we will use experimental research results to develop simulation and performance models. The performance and simulation models are linked since local EDPs are needed to assess the structural performance of a beam-column joint. Here these model pairs will be developed for two categories of simulation models: simplified models, needed for immediate use by the profession, and more sophisticated models, needed for immediate use by researchers and future use by industry.

The initial research efforts (Tasks 1 and 2) will focus on OpenSees model formulations, including formulations by Lowes et al. and Deierlein et al. Experimental results from the PEER research on existing beam-column joints will be used to calibrate and validate the models (Walker 2002, Alire 2003). Similar efforts will develop simpler models for immediate use by industry.

After completing Tasks 1 and 2, the simulation models will be used to develop performance models for evaluating the component performance. Research results indicate that joint damage is related more closely to local EDPs, such as joint shear strain, rather than global measures, such as drift. Since the joint shear strain depends on the formulation of the simulation model, a performance model must be developed for each simulation model (Task 3). This development will occur in three phases, as illustrated in Figure 1:

  1. Develop fragility curves using measurements (Alire 2003)
  2. Simulate specimens using model of interest
  3. Use the results of Phase (a) and (b) to develop simulation-based performance models

Finally, the research results will be validated using other experimental data which differ from those used for model development. The researchers will coordinate their efforts with Professor Lowes in this validation and work to provide results to her in a timely manner to facilitate use of these models for building simulation.


Figure 1.

Larger View

Guidelines for the use of each of pair of models (i.e., simulation and performance model pair) are needed for the practicing engineering community. To facilitate this, a concise set of model guidelines including the necessary equations, recommended values for the impute parameters, and numerical examples, will be prepared. The strengths and limitations of each model will also be outlined. The research team will work with the PEER partners within the BIP program to facilitate the transfer of the results from research-to-practice.
Brief description of past year’s accomplishments and more detail on expected Year 6 accomplishments

The Year 5 research effort has used the experimental research results to develop tools that will support the proposed Year 6 effort. These accomplishments include:

  1. Structural performance states for joints without transverse reinforcement.
  2. Development of performance models for three structural performance states, including initial cracking, initial spalling, and severe spalling. The performance models use the measured joint shear strain and are presented using fragility curves and mean values.
  3. Development of a joint shear stress-strain constitutive model for joints without transverse reinforcement.
The model accounts for strength and stiffness degradation due to cycling. Strength and stiffness degradation in the unloading and reloading portion of the curve (or pinching) is also modeled. The model has been developed using the PEER experimental research results and validated using results from other related research efforts.
Other similar work being conducted within and outside PEER and how this project differs

The research results from this project will provide information for PEER researchers, non-PEER researchers, and structural engineering professionals. The research effort complements and is needed for the building simulation effort by Lowes. The models developed can be modified by others studying joints with transverse reinforcement (e.g, Jim Lafave U.of Illinois).

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

Tools for the seismic evaluation of these types of structures must be verified using experimental data. A large number of the joints in older building frames are eccentric, with the centerlines of the beams and columns offset from each other and since no studies have focused on such joints without transverse reinforcement, their performance remains relatively unknown. In addition, since ACI 318-02 penalizes eccentric joints by assigning to them a small area, the relative joint shear stress is increased which may increase the perceived need for retrofit. Research into the performance of eccentric beam-column joints is needed. The results can be used to modify the proposed simulation and performance models and to provide the profession with models that better meet their needs for seismic assessment.

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

The researchers are cognizant that their research results are not currently being used by industry. This research proposal was developed, in part, to remedy this deficiency.

Expected milestones
  • Development of simulation tools for immediate use by industry
  • Validation of more advanced simulation tools.
  • Development of performance models appropriate for each type of simulation model, including acceptance criteria.
  • Development of guidelines for the use of each pair of simulation and performance models.
  • Practical guidelines for the use of each pair of simulation and performance models.
  • Fragility curves for OpenSees simulation models for B/C joints