Projects
- Rebar Depth Effect on Bridge Deck GPR Assessment
This project will use simulations of GPR bridge deck investigations to develop predictive models for the influence of rebar depth on the amplitude of rebar response. The model predictions will be compared to field data, and used to develop algorithms to correct for the rebar depth effect.
- System Identification and Structural Diagnostics
- 2D & 2.5D versus 3D FDTD GPR Modeling of Bridge Decks
Computational modeling is beneficial in the preparation for non-destructive wave based sensing.
- Characterization of a GPR Antenna for Excitation of a 2D FDTD Model of Reinforced Bridge Decks
This research uses experimental results from an uncharacterized GPR antenna to determine the signals required to excite a 2D FDTD model. The 2D FDTD virtual sensor will consider antenna directivity and its use will produce realistic simulation results.
- Civil Infrastructure Monitoring for Change Detection using an Informational Entropy Model Framework
The objective of this project is to formulate an imaging-based informational entropy model framework for the detection and estimation of certain classes of changes in the infrastructure in urban settings. Taking signals as input, the idea is to process those using entropic measures to detect changes by comparison with a previous state.
- Damage Localization Accuracy as a Function of s in the Complex Plane
This project explores whether strategic s-selection can refine results from the dynamic damage localization vector (DDLV) strategy. Part of the DDLV strategy involves comparing the transfer function of a potentially damaged system to the transfer function of the system's undamaged state.
- Determination of Foundation Depths
This research, conducted in collaboration with Boston Groundwater Trust, investigates the use of electromagnetic waves to determine the foundation piling cutoff depth.
- Efficient Computational Methods and Sparse Sampling for Dominant Mode Extraction in Distributed Structures
Empirical bases, based on the proper orthogonal decomposition (POD) provide efficient, low dimensional representations of distributed. Very high data dimensionality (O(10^6) – O(10^9)) is commonplace in such applications as fluid dynamics and aero-elasticity, making the processing of long data streams computationally expensive.
- Fusion of Electromagnetic & Mechanical Wave Data for Concrete Structure Diagnostics
The research objective is to develop fusion models that are generally applicable to non-destructive concrete structure diagnostics. The approach will be general and will be demonstrated on two pervasive and significant defects found in concrete structures: delaminations and voids.
- Low Order Dynamic Galerkin Models on Nonlinear Manifolds: A Deformable Modes Approach
Reduced order models are essential for effective control design and even shape optimization in complex distributed systems, such as complex flexible structures, fluid dynamic systems and aero-elastic fluid-structure interactions. Low order Galerkin models, based on proper orthogonal decomposition (POD) modes, are a natural choice.
- Tunnel Detection Project
The objective of this research, conducted with TransTech, Inc., is to sense the presence of underground tunnels using electrical impedance tomography.
- Using Simulations to Investigate the Performance of Residual-based Techniques for Damage Detection
One approach that has been used to detect and locate damage is an Autoregressive with Exogenous Input (ARX) model. It has been shown analytically that when inputs and input locations are unknown (as is frequently the case in structural health monitoring) the ARX predictor yields results which are filtered versions of open-loop predictions.
