Research Interests:

General: Computational mechanics

Particular: Wave propagation, scattering and radiation phenomena; structural dynamics;
                Inverse problems in engineering;
                Earthquake engineering (ground motion characterization);
                Acoustics and structural acoustics; Fluid-structure interaction;
                Computational biomechanics;
                Numerical methods (FEM, BEM and SGBEM, fictitious domain, etc);
                Scientific computing (parallel and distributed model-based simulations);
                Large-scale visualization

Projects (current):

5. CAREER: Towards near-real-time site characterization: Advanced computational methods and NEES-based validation experiments
PI: L.F. Kallivokas; Source: NSF

The aim of the project is multifold: a) to develop the capability for the site-specific rapid imaging of the skeletal properties of a soil mass by coupling in-situ non-invasive experimental and computational methods; b) to develop the computational framework and experimental protocols that will allow near-real-time profiling of large sites, including in-situ adjustments in field arrays to optimize inversion procedures; c) to seek to validate the proposed approach by profiling selected sites of the Imperial Valley in California; and d) to migrate to the educational curriculum the experimental and computational techniques to be developed, aiming at the training of next-generation engineers to the state-of-the-art in an area of grave importance to seismic hazard mitigation efforts.

4. ITR-Collaborative Research: High-fidelity, high-resolution earthquake modeling: Dynamic rupture, blind deconvolution imaging, and ultrascale computing
PIs: J. Bielak (CMU), O. Ghattas (CMU), S. Day (SDSU), L.F. Kallivokas; Source: NSF

Recent advances in (1) fault-rupture modeling, (2) forward modeling methods for earthquake ground motion in large basins, and (3) partial differential equations (PDE)-constrained optimization methods, combined with the increasing availability of earthquake records from new strong-motion and broadband sensor networks, make it possible for the first time to create fully realistic three-dimensional inversion-based models of complex basin geology and earthquake sources, and to apply such capability to model and forecast strong ground motion during earthquakes.

The goal of the project is to develop this capability. This is a multi-institutional project involving San Diego State University (S. Day) and Carnegie Mellon University (J. Bielak, O. Ghattas). The particular focus at The University of Texas at Austin is the exploitation of canonical grid methods (fictitious domain, mortar element methods) for the modeling of dynamic seismic fault ruptures.

3. Optimal control of distributed parameter systems
PI: L.F. Kallivokas; Source: none

Borrowing from band-gap ideas, we explore the feasibility of reducing the surface and near-surface motion that propagating seismic (elastic) waves have on structures, through alterations in the soil's material profile close to the structure. The problem is cast as a distributed parameter control where we seek to uncover periodic material structures over a near-site control region that will allow the motion's amplitude reduction.

2. Determination of fatigue damage in stay cables
PIs: S. Wood, K. Frank, L.F. Kallivokas; Source: TxDOT

Both of the cable-stayed bridges in Texas have experienced large-amplitude cable vibrations during the past ten years. TxDOT has modified the bridges by adding cable restrainers and dampers to reduce the likelihood of large-amplitude cable vibrations in the future, but the extent of fatigue damage is not known. The objective of this research project is to determine the likely locations of fatigue damage and estimate the extent of the problem. To date, seven stay cable specimens have been tested in the laboratory and acceleration data have been collected from the Fred Hartman Bridge.

Our goal is to develop analytical predictive models of the stay cables, which will be used to relate the observed response of the laboratory specimens to the stays in the field.

1. Vessel impact on bridges
PIs: L. Manuel, L.F. Kallivokas, E. Williamson; Source: TxDOT

Our interest here is in the estimation of vessel-to-pier impact forces using finite element modeling. Parametric studies using data on various relevant bridges and vessels will be carried out. The focus will be primarily on representative configurations of ships and barges commonly encountered in the Texas waterways, and in modeling the impact for a selected set of piers for the most critical bridges.