Structural Engineer Stantec
Modeling Seismic Degradation and Axial Collapse of Reinforced Concrete Columns Using Nonlinear Continuum Finite Elements
Abstract: Reinforced concrete columns are critical to seismic stability; however, experimental data describing column behavior up to axial collapse remain limited, particularly for different loading histories and degradation modes. This study presents practical recommendations for nonlinear continuum finite element (FE) modeling of reinforced concrete columns subjected to cyclic and monotonic seismic loading up to axial failure. Three-dimensional FE models were developed and calibrated using results from twenty-one full-scale column tests exhibiting shear-, flexure–shear-, and flexure-critical degradation. Several datasets were obtained from the DesignSafe Data Depot to support model calibration and validation.
The modeling framework evaluates the influence of key material and cyclic parameters, including concrete softening behavior, unloading response, plastic flow direction, and reinforcement bond–slip assumptions. Model accuracy was assessed using peak lateral strength and drift-based performance metrics associated with lateral and axial degradation. The calibrated models showed good agreement with experimental responses across a broad range of axial loads, shear stresses, and reinforcement configurations, with greater sensitivity observed for short columns and high axial-load conditions. Comparisons with ASCE/SEI 41-17 and ACI 318-19 indicate that code-based shear capacity provisions are generally conservative relative to calibrated FE results.
Although analyses were conducted using a specialized concrete FE platform, the workflow was structured to align with SimCenter performance-based methodologies, incorporating quoFEM-style uncertainty and sensitivity assessment concepts. The proposed recommendations are intended to support future parametric studies and improve seismic performance assessment of reinforced concrete columns approaching collapse.
Assistant Professor University of Minnesota, Twin Cities
Nonlinear Spring Calibration for Reinforced Concrete Column-Foundation Connection Bar Group Breakout Failure
Co-Author: Jorge Archbold (Universidad del Norte)
Abstract: Concrete breakout is a bar group failure mode that can govern the strength and deformation capacity of reinforced concrete (RC) column-foundation connections. Bar group breakout can govern even when development length requirements are met. Although recent tests have documented this brittle failure mode, these mechanisms are rarely modeled in nonlinear structural simulations. To address this, ACI 318-25 introduced new bar group breakout provisions (Section 25.4.11), including the use of distributed shear reinforcement to improve strength and ductility. While these provisions were calibrated at the component level, their effect on overall building collapse performance remains unquantified.
This study documents the development and calibration of nonlinear spring models to represent the moment-rotation behavior of breakout-prone column-foundation connections. Calibration is performed using experimental force–drift data and the quoFEM application from the NHERI SimCenter, which employs Bayesian parameter estimation via Transitional Markov Chain Monte Carlo (TMCMC). Three OpenSees uniaxial material models—Hysteretic, and Modified Ibarra-Medina-Krawinkler—are evaluated for their ability to reproduce degrading hysteretic behavior.
The calibrated springs may be integrated into building models to assess structural collapse performance per the FEMA P695 methodology, evaluating whether updated designs meet the life safety targets implied by modern codes (e.g., <1% collapse probability in 50 years). This work demonstrates the value of NHERI-supported computational tools in bridging component-level testing and system-level performance evaluation for emerging code provisions.
Postdoctoral Scholar University of Washington
Integrated Computational Framework for Seismic Load Paths and Functional Recovery in Cast-in-Place Diaphragms with Openings
Co-Authors: Travis Thostad (University of Washington) and Dawn E. Lehman (University of Washington)
Abstract: Cast-in-place (CIP) reinforced concrete (RC) diaphragms play a critical role in transferring inertial forces in buildings yet are often represented using simplified assumptions in seismic design. Recent shake-table tests and post-earthquake field observations indicate that neglecting the interaction between diaphragm openings and vertical lateral-force-resisting elements can lead to severe and unexpected damage, compromising structural safety and causing significant delays in post-earthquake functional recovery.
To address this challenge, an integrated computational-experimental research program is currently underway. The presentation details the development of a modeling approach for CIP diaphragms, calibrated using data from sub-assemblage tests of critical structural components and validated using both global and local response data from a five-story RC shake-table specimen that was tested in 2012, available through the DesignSafe Data Depot.
A sub-modeling strategy is employed to bridge global structural response and local component behavior. Displacement boundary conditions and stress histories extracted from global building simulations are used to drive high-resolution analyses of isolated diaphragm segments, utilizing DesignSafe-CI high-performance computing (HPC) resources at TACC. These digital sub-models will be validated against physical panel experiments, enabling detailed interrogation of complex behaviors - such as shear transfer around openings - that are computationally prohibitive to resolve within global models alone. This framework establishes a feedback loop in which global simulations provide realistic boundary conditions for component-level investigation, while component response data inform and refine global predictions. The ultimate objective is a design-oriented framework that accounts for opening-induced demands, ensuring load-path integrity and advancing the post-earthquake functional recovery of buildings.
PhD Student University of Missouri–Kansas City
Three-Dimensional Nonlinear Simulation of Soil–Structure Interaction for Low- and High-Rise Reinforced Concrete Buildings Under Earthquake Loading
Co-Author: Mohanad Abdulazeez (University of Missouri–Kansas City)
Abstract: This study investigates how soil–structure interaction (SSI) influences the seismic response of low-rise and high-rise reinforced concrete buildings founded on soft and hard soils subjected to earthquake loading. A computational investigation was performed using OpenSees to simulate three-dimensional building systems of distinct heights, representing low-rise (5-story) and high-rise (40-story) structures, capturing material nonlinearity, dynamic response characteristics, and soil-dependent amplification effects. Ground motion records, including representative historical earthquakes, were applied to evaluate differences in global response, interstory drift, and structural demands, while probabilistic post-processing using QuoFEM assessed damage indices and potential failure mechanisms. This study undertakes a systematic parametric evaluation of the coupled effects of soil stiffness and structural height on seismic demand allocation, damage accumulation mechanisms, and global structural vulnerability. Emphasis is placed on identifying governing response quantities responsible for damage amplification in soft soil environments relative to those in stiff soil sites, as well as on characterizing the height-dependent sensitivity of reinforced concrete building systems to soil–structure interaction phenomena. The study delivers a scalable and transferable computational framework for quantifying soil–structure interaction effects, enabling enhanced seismic risk evaluation and supporting performance-based design and resilience-oriented planning for buildings in earthquake-prone regions. The results indicate that tall buildings on soft soils exhibit significantly higher damage, whereas low-rise buildings show minimal damage, suggesting that low-rise construction on soft soils and high-rise construction on stiff soils is preferable.
PhD Student University of Missouri-Kansas City
Stochastic Buckling Reliability of Thin-Walled Tubular Sections: An AI-Enhanced Workflow using quoFEM and LS-DYNA
Co-Author: Mohanad Abdulazeez (University of Missouri–Kansas City)
Abstract: The axial stability of large-diameter hollow structural sections (HSS) is paramount for the resilience of critical energy infrastructure like wind turbine towers and offshore monopiles under extreme wind hazards. Current design practice typically accounts for imperfection sensitivity through conservative, deterministic knockdown factors, often without explicitly quantifying the associated uncertainty or collapse probability, potentially leading to inefficient material use. This research presents a novel computational workflow that moves toward a rigorous, reliability-based design paradigm by explicitly quantifying the stochastic buckling strength of thin-walled columns. A high-fidelity finite element model is developed in LS-DYNA to capture nonlinear local buckling, post-buckling strength degradation, progressive crushing, and energy absorption. The model is calibrated and validated against experimental monotonic axial compression tests on bare steel HSS columns. This validated model is integrated with the NHERI SimCenter tool quoFEM to generate stochastic imperfection fields representing realistic manufacturing variances. An active learning scheme using Gaussian Process surrogates is implemented within quoFEM to enable effective exploration of the imperfection space. This AI-enhanced framework efficiently maps the imperfection parameter space to axial performance metrics, including ultimate strength, post-buckling load retention, and energy absorption. The outcome is a probabilistic characterization of imperfection sensitivity, providing direct inputs for the development of risk-informed design criteria for hazard-critical renewable energy infrastructure.
Postdoctoral Scholar University of Pavia
A novel macroelement-based strategy for modeling masonry structures
Co-Authors: Gabriele Guerrini (University of Pavia), Alessandro Galasco (University of Pavia), and Andrea Penna (University of Pavia)
Abstract: Unreinforced masonry (URM) buildings are among the most vulnerable structural typologies under seismic actions, mainly due to inadequate seismic design, poor structural detailing, and unfavorable mechanical properties of masonry constituents. Equivalent-frame modeling with nonlinear macroelements is widely adopted for multiple and large-scale nonlinear seismic analyses of URM structures, as it balances computational time with accuracy of results compared to more refined micro-modeling techniques.
This contribution presents a two-node three-dimensional macroelement that extends conventional equivalent-frame formulations by introducing an efficient representation of the coupled in-plane and out-of-plane axial-flexural response under lateral loading, enabling improved seismic assessment of URM elements. The versatility of the proposed formulation also allows the explicit modeling of lumped and distributed reinforcement within the macroelement at limited computational cost, thereby extending its assessment and design capabilities to a wide range of strengthened or reinforced structural typologies.
The proposed macroelement is validated against experimental results on strengthened and unstrengthened structural configurations, demonstrating accurate reproduction of lateral strength, stiffness degradation, hysteretic response, and displacement capacity.