Relating EDPs in RC Bridges to Damage and Decision Metrics - 2452005

Project goals and objectives

The overall goal of this project is to relate Engineering Demand Parameters (EDPs) to Damage Measures (DMs) and Decision Variables (DVs) for reinforced concrete bridges. The specific objectives are to

1. Evaluate and calibrate the OpenSees column modeling strategies using the PEER Structural Performance Database, and for complex loading conditions, including varying axial loads, bi-directional loading and shaking-table experiments.
2. Calibrate and evaluate the benefits/drawbacks of implementing new material models for the cyclic response of reinforcing bars (Mohl-Kunnath) and concrete (Mitra-Berry- Lowes).
3. Provide simulation support for PEER bar buckling experiments (Lehman-Stanton)
4. Coordinate WSDOT, FHWA and PEER-sponsored modeling (Mahin, Billington) on the development of post-tensioned systems for bridge columns.
5. Assist Stojadinovic on the development of methodologies to estimate a general set of EDPs, DMs and DVs for reinforced concrete bridges.
6. Continue to maintain and (if possible) expand the Structural Performance Database, which provides column data for nearly 500 columns.

Role of this project in supporting PEER's mission (vision)

The successful application of the PEER methodology depends (among other things) on the development and calibration of tools and methodologies to predict the force-displacement response, damage development and damage consequences for key structural elements. Column damage is one of the most common and most important types of damage in bridges and buildings.

Methodology employed

The research description is organized into six sections, according to the project objectives listed in the preceding section.

Column Modeling

The spread-plasticity and lumped-plasticity column models have been evaluated and calibrated using the detailed response measurements from Lehman-Moehle column tests, and using the larger set of column tests documented in the PEER Structural Performance Database (Berry- Eberhard). During Year 9, the modeling strategies will be evaluated and calibrated using cyclic tests of columns with bi-directional loading and/or variable axial loads (e.g., Xiao), and tests of column on the shaking table (e.g., Mahin and Sakai).

On the practical side, Berry, Lehman and Lowes will submit a manuscript on the plastic-hinge modeling of bridge columns in April, 2006. Elwood (UBC) and Eberhard wrote a PEER research digest, which evaluates the current FEMA recommendations for estimating column effective stiffness and which recommends a new methodology. A key plot from the digest appears below.

Evaluation and Calibration of New Material Models

The accuracy of estimates of column behavior will be improved by evaluating and calibrating two new material models. The Mohl-Kunnath reinforcing bar model, which was recently added to OpenSees, will be calibrated (using the performance database) to improve force-deflection estimates (especially stiffness and strength degradation) and predictions of bar buckling. The Mitra-Lowes concrete model (concrete04) will be modified to improve the accuracy of the reloading branches, which will likely lead to better estimates of residual displacements. These improved estimates are critical to the implementation of the PEER methodology for bridges, in which residual displacement is a key damage measure. This measure is particularly important in evaluating the merits of post-tensioned column construction, such as Billington and Mahin are doing.

Simulation Support for Bar-Buckling Study

The bar buckling study (Lehman and Stanton) needs to be closely coordinated with the development of column modeling recommendations. During Year 9, we will conduct analyses in support of this project, and will ensure that the lessons learned are incorporated into the modeling recommendations.

Post-Tensioned Bridge Columns in Seismic Regions

As part of a WSDOT-sponsored project, Stanton and I are developing precast bent systems that can be constructed rapidly and that have adequate seismic performance. Self-righting systems are an important component of our work. In communicating our results with WSDOT, we are adopting many of the tools, language and concepts of the PEER performance-based methodology.

During Year 9 (and beyond, if funded), we will coordinate these activities with those of Mahin and Billington on self-righting reinforced concrete bridge systems, and will take advantage of synergies between the projects.

General Set of Damage Measures and Decision Variables

So far, much progress has been made in modeling the flexural performance of reinforced concrete columns in bridges. Much less attention has been paid to other types of damage (e.g., residual displacements, unseating, joints) that have important consequences for bridges. We will assist Professor Stojadinovic in developing a preliminary set of damage measures and relationships to decision variables that are important to bridges (e.g., closure, downtime, repair cost). During Year 9, we will focus on assembling and packaging previous work (e.g., Lehman/Stanton/Lowes, Porter, DesRoches), developing an expert consensus where little data is available, and identifying strategies for improving the estimates of EDPs and DM-DV relationships.

Maintain and Expand Performance Database

As new data becomes available, the database is being updated and expanded.

Brief Description of previous year's achievements, with emphasis on accomplishments during last year (Year 8)

During Year 8, we developed and calibrated strategies for modeling column force-displacement response (related to EDMs) and damage progression (related to DMs) using both distributed- plasticity and lumped-plasticity modeling strategies. These modeling strategies have been implemented by a series of other PEER project, including the bridge testbed (Mackie and Stojadinovic)

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

To my knowledge, no other work has been done on systematically evaluating the accuracy of OpenSees modeling strategies using column data, either within or outside of PEER.

This project needs to be coordinated with many other current PEER projects, including:

• - Lehman and Stanton experimental study on bar buckling
• - Kunnath project development of bar buckling model
• - Lowes modification of Concrete04
• - Mahin and Billington projects on self-righting bridge bent systems
• - Stojadinovic project on general EDP-DM-DV relationships.

This project will also take advantage of previous PEER projects, including Lehman, Stanton and Lowes studies on joint vulnerability, Porter study on DM-DV relationships, and column testing projects (e.g., Xiao).

Expected milestones & deliverables

• April 1st:, 2006: Elwood and Eberhard submitted PEER research digest
• April 15th, 2006: Berry, working with Lehman and Lowes, will submit manuscript describing the development of a new plastic-hinge model and its application to estimating likelihood of column damage.
• June 30th, 2006: Participate (with Stojadinovic) in workshop with Washington State DOT on DM-DV relationships.
• Sept 30th, 2006: Submit formal PEER research report on modeling of RC columns with OpenSees

Member company benefits

The development of a general set of EDP-DM-DV relationships for bridges is critical to applying the PEER methodology in practice, and in demonstrating its value to Caltrans and other bridge agencies.

The use of the PEER methodology in discussions with WSDOT and Caltrans will demonstrate the practical implementation of the methodology in developing and evaluating new systems.