Guidelines for Nonlinear Analysis of Bridge Structures - 9A01

Project Title—ID Number	Guidelines for Nonlinear Analysis of Bridge Structures - 9A01
Start/End Dates	1/15/06 – 1/15/07
Funding Source	PEER-Caltrans
Project Leader (boldface) and Other Team Members	Bozidar Stojadinovic (UCB/F), Kevin Mackie (UCB/PD), Vesna Majstorovic (UCB/GS), Ady Aviramtraubita (UCB/GS)

Project goals and objectives

1. Ensure that accurate nonlinear modeling techniques are employed by Caltrans.
2. Ensure that PEER researchers realistically model typical Caltrans bridge details.
3. Identify incompatibilities or inconsistencies between SAP2000 and OPENSEES.
4. Identify bridge components that require special modeling consideration.
5. Establish criteria for when specific bridge components require nonlinear characterization
as well as the level of sophistication of that nonlinear characterization.
6. Develop summary guidelines for nonlinear modeling of bridges.

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

The work in this project may be viewed as the transition-to-practice of the projects on probabilistic performance-based evaluations of bridges conducted within the PEER Bridge Thrust. The project in the Bridge Thrust are grouped into three groups: (1) the predominantly geotechnical ones, focusing on assessing liquefaction and soil spreading hazard; (2) the predominantly structural ones, focusing on fragility assessment of bridge components and development of new design concepts, such as application of new materials and devices to reduce residual displacement, to improve bridge seismic resistance; and (3) the system-related ones, where complex models of a set of test-bed bridges are developed to asses the probability distribution of decision variables, such as the repair cost or the ability to carry traffic load after an earthquake, and to evaluate the quality of combined soil-structure models to model soil- structure interaction.

Methodology employed

The project will be performed in three stages.

In Stage 1, a comparison of SAP2000 and OPENSEES will be made using 4 bridge models previously developed by Tsai (2003). Each bridge will be run longitudinal only, transverse only, and longitudinal and transverse combined. For transverse loading SDC recommendations for abutment modeling will be used. For longitudinal loading abutment models developed by Tsai will be used. Caltrans will provide the data required to develop the OPENSEES and SAP2000 models, as well as the existing SAP2000 (if any) at the start of the project.

For each bridge, comparisons will be made using approximately 3 time-series, each scaled to a low, moderate, and high level of excitation. (i.e. linear, ductilities of 2-3, and ductilities of 4-6). The ground motions will be chosen using the PEER ground motion database in consultation between the PIs and Caltrans engineers within the first half of Stage 1 of the project.

The primary objective of Stage 1 is to identify and resolve inconsistencies between SAP2000 and OPENSEES. Issues requiring parametric study will be identified.

In Stage 2, parametric analysis will be performed to quantify the consequences of various modeling choices through determination of median and sigma structural response resulting from excitation by a suite of input motion (approximately 30 records). The ground motions will be chosen using the PEER ground motion database in consultation between the PIs and Caltrans engineers at the beginning of Stage 2 of the project. Modeling choices will be evaluated in consultation between the PI and Caltrans engineers during Stage 1 of the project so as to arrive at a defined set of alternatives early on in Stage 2 of the project.

Stage 3 consists of developing guidelines as described in Section 2 (above) utilizing the results of the parametric study in Stage 2 (as well as other sources selected by the PI). These guidelines will be developed in consultation with Caltrans engineers.

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

This is the first year for this project.

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

It is anticipated that the models developed for this study (both SAP2000 based and OPENSEES based) will be of potential value to investigators studying issues of time-series selection and scaling. These models will be made available if requested.

Two NEES Small Group awards have been made in the bridge area. One project focuses on the seismic simulation and design of bridge columns under combined actions and implications on the system response. This experimentally-based project aims to produce fragility relations for columns under combined loading, analysis guidelines and design methods for combined bridge column loading. However, it is not clear if the columns tested will be similar to the typical Caltrans columns. The other project focuses on the seismic performance of bridge systems with conventional and innovative materials. This project comprises tests of a full-scale abutment model and tests of a four-span bridge models on a shaking table. The expected outcomes of the project are better understanding of the response of bridge systems, with and without soil- structure interaction, development of wireless sensors and numerical modeling using OpenSees. Again, it is not clear if the examined bridge structures will conform to Caltrans SDC. Thus, while these projects are certainly interesting, it is not clear if they will apply to typical Caltrans bridges directly.

Work on analysis and seismic evaluation of bridges is ongoing within the MAE and MCEER centers. The MAE Center is focused on typical East Coast bridges. While the approach taken by the MAE Center is quite similar to the PEER Center approach, the bridge details are so different from typical California bridges that only a small fraction of their findings may apply. The MCEER Center is focused on developing REDARS. Caltrans is already participating in this effort.

Work conducted within the NCHRP 12-49 project on developing a comprehensive specification for the seismic design of bridges resulted in an NCHRP Report 472 which contains the finding of the project and the background for the AASHTO Guide Specification. The findings of this report that pertain to bridge modeling and, further, to design philosophy will be taken into account in this project.

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

None just yet.

Expected milestones & deliverables

Stage 1 SAP2000 – OPENSEES comparison report: Following comparisons of SAP2000 and OPENSEES bridge models, the PI will prepare a short report that provides guidelines for nonlinear bridge modeling using SAP. This report will also identify unresolved issues that will be the focus of parametric study in Stage 2.

Stage 2 data report: Following parametric evaluation of various bridge models and modeling options, a data report will be prepared documenting median and dispersion (sigma) values of displacement demand for the various models considered in the parametric study. A simple description of the models considered will also be provided.

Final Report: A final report will be prepared that includes guidelines for (1) determining when linear methods are sufficiently accurate to estimate structural demands, (2) determining what components require nonlinear characterization (when nonlinear analysis is required), and (3) how to model nonlinear bridge components.

Member company benefits

The objective of this project is the development of practical guidelines for non-linear analysis of bridge structures. The goal is to help practicing engineers implement nonlinear methods for bridge design and analysis in their every-day practice. The project is viewed as a collaborative effort between university researchers and practicing bridge designers. As such, it is hoped that these guidelines will be readily implemented in practice.