Seismic Response of Reinforced Concrete Walls

Summary

This project includes two critical problem-focused tasks in Performance Based Seismic Engineering (PBSE): (1) the evaluation of reinforced concrete wall models used in high-fidelity nonlinear dynamic analysis, paralleled by development of a new wall model for use by practitioners using commercial software, and (2) an experimental investigation of the effect of reinforced/concrete wall slenderness and tension on ductility and capacity, all with the goal of developing new codes and standards provisions .

Description

Objective: The project includes two critical tasks identified in Performance Based Seismic Engineering (PBSE): (1) use of simplified reinforced concrete wall models in high-fidelity nonlinear dynamic analysis that have no robust experimental data validation - to be solved by thorough experimental data verification and development of improved models; and (2) investigation of global wall bucking in the 2010 Chile earthquake designed using comparable US practice, to be solved by an experimental investigation of the effect of wall slenderness and tension pre-loading on ductility and capacity.              

What is the new technical idea? There are two separate problems addressed in this project:  

Reinforced Concrete Wall Models for Seismic Response  

This task will connect and overlap with the three constituent communities involved in the nonlinear seismic analysis of reinforced concrete walls: the design community, commercial software vendors, and academic researchers. Each of the constituent groups typically operates primarily within its own domain (aside from software vendors catering to the design community). The new technical idea is to evaluate existing models implemented by software vendors as used by the design community, using experimental data and observations in the hands of the academic researcher, as well as to evaluate the models used by researchers that are not implemented by the software vendors or used by the design community because of their inherent complexities. As noted by the Principal Investigator during the Chile earthquake reconnaissance, the failure mode of many reinforced concrete walls would likely not be predicted using current commercial and research finite element software. The best properties of models used by both communities will be combined to develop a new wall finite element formulation. It is anticipated that the new wall model may be fiber-based, but will include interaction with membrane shear. The goal is to develop a wall model targeted at the design community that is robust and readily implemented by the software vendor, is computationally efficient, and captures the behavior observed by the academic and research communities.  

Performance of Slender Reinforced Concrete Walls  

The 2010 Chile earthquake highlighted shortcomings in the design configuration of reinforced concrete structural walls. Many walls that were designed essentially in accordance with customary US practice, except for confinement detailing, exhibited out-of-plane buckling due to slenderness. This behavior has become a major point of discussion in recent months in ongoing NEHRP Consutlant Joint Venture (NCJV) contracted work on two separate task orders related to wall design and Chile earthquake implications. The consensus of the experts on those efforts is that this observed behavior has serious negative implications for US design practice, and is not well understood. It is postulated that large tension excursions contributed to the observed buckling behavior. Current US model building codes do not have limits on wall slenderness, and with increasing architectural demands in the absence of these limits, the poor behavior of walls as observed in Chile could be realized in the US.   

This proposed task addresses the behavior of slender walls, with an ultimate impact on US building code requirements (IBC [1], ACI [2]). The output of this task will be based on new experimental data obtained through contracted structural testing, and nonlinear analysis. It is proposed that a significant portion of the work, including the analytical and experimental parts of the task,  be carried out by a graduate SCEP student, Travis Welt, under the direction of Doctors Dragovich and McCabe. The project will form the basis of Mr. Welt's PhD thesis.              

What is the research plan?  

Reinforced Concrete Wall Models for Seismic Response  

This task will be begin with collecting existing experimental data related to reinforced concrete wall cyclic response. An extensive literature survey will be performed, and personal contacts in academia who have experimental databases [3], [4] will be contacted for collaboration. Finite element models will be developed to predict the response of the experimental data using available commercial nonlinear finite element software. This phase will be extensive, requiring the development of many nonlinear finite element models using commercial software and "high-fidelity" models using LS-DYNA. The effectiveness of the finite element models in predicting the response of structural walls to earthquake ground motions will be quantified. Subsequently, the research will identify and implement "research" wall element models (ones not used in industry) that accurately predict observed response with high fidelity. The results of the commercial and research model predictions will be compared to assess their effectiveness, and conditions when the various models provide satisfactory prediction of observed response will be identified. Based on this comparison, a new finite element formulation for walls will be developed specifically for seismic applications, using the best properties of the models evaluated. A new DWNN (Dynamic Wall NIST NEHRP) model that is compatible for inclusion in general purpose finite element software packages will be targeted for use by earthquake engineering practitionerse finite element software packagpackages ral t the Washington Wall Ladies. At the end of the task, the outcome will be numerical models that are readily usable by earthquake engineering practitioners, are readily implemented by software vendors, and have been shown to provide strong linkage between observation and analysis. It is anticipated that future efforts in support of PBSE will be undertaken to address other aspects of the PBSE issue. The task will engage a NCJV peer-review panel to foster broad acceptance.  

In addition, a collaborative agreement has been established with Professors Laura Lowes and Dawn Lehman from the University of Washington to review their extensive test results and large test database, as well as to work together on analysis and building code issues. They have a large NSF/NEESR wall project now and extensive experience in this area. NSF has provided funding to support a UW doctoral student in residence for several months in Spring 2012.  

Performance of Slender Reinforced Concrete Walls  

The task goals will be achieved through a combination of numerical simulation and an experimental program. The major phases of this project are literature review, experimental design and prediction, the experimental program, and refinement of predictions and code recommendations. The concept of the experimental program is to test approximately 20 shear wall panels designed to fail in flexure. The testing program configuration will be developed to be simple and cost-effective, and carried out through an external structural testing laboratory that has demonstrated expertise in earthquake engineering research and large-scale structural tests. The testing award will be made through a competitive RFP process. The major variables of the testing program will be: wall slenderness (wall height-to-thickness ratio), absolute wall thickness, gravity stress as a function of f'c, wall confinement, and loading protocol. The loading protocol will be developed to investigate the effect of significant pre-loading in tension, and its impact on wall buckling. Upon completion of the experimental program and analysis, building code recommendations will be developed.  

The project will be carried out by the co-PI Travis Welt (SCEP), with the work forming the basis of his Ph.D. dissertation at the University of Illinois. He has complete mentoring support from his thesis advisor there, as well as in-house mentorship from Doctors Dragovich and McCabe. Mr. Welt will be completing the initial literature review and experimental design summer of 2011.

[1]International Building Code, International Code Council, Washington, D.C. 2009

[2] Building Code Requirements for Structural Concrete, ACI 318-08, American Concrete Institute, Framingham, MI, 2008

[3] Wallace, John, Modeling Issues for Tall Reinforced Concrete Core Wall Buildings, Structural Design of Tall and Special Buildings, Dec 2007.

[4] Orakcal K, Wallace J, Flexural modeling of reinforced concrete walls - Experimental verification, ACI Structural Journal,  vol. 103, issue 2, pp 196-206, Mar-Apr 2006.

Major Accomplishments

Recent Results:  None - this is a new project.              

Standards and Codes:   

Improved Reinforced Concrete Wall Models for Seismic Response  

The results of the task will be formulated to provide guidance in building code provisions  on acceptable methods of analysis of reinforced concrete walls, for inclusion in future versions of ASCE 41[1], ASCE 7[2] and ACI 318  

Performance of Slender Reinforced Concrete Walls

The results of the task will be formulated to provide guidance in building code format on slenderness and tension controlled actions of reinforced concrete walls, for inclusion in future versions of IBC, ASCE 7 and ACI 318. [1] Seismic Rehabilitation of Existing Buildings, ASCE 41-06, American Society of Civil Engineers, Reston, VA 2006 [2] Minimum Design Loads for Buildings and Other Structures, ASCE 7-10, American Society of Civil Engineers, Reston, VA 2010