Research Overview
The applied structural dynamics lab (ASDL) analyzes the dynamic response of civil infrastructure such as bridges, power plant structures, and buildings through the rigorous dynamic analysis and advanced real-time testing. The major research interests of ASDL include, but not limited to, structural vibration control, seismic performance evaluation of structures, nonlinear dynamic analysis, structural health monitoring, and real-time hybrid simulation (RTHS).
Current Research
Multi-Axis Real-Time Hybrid Simulation
Multi-axis real-time hybrid simulation (RTHS) is an advanced experimental technique in civil engineering that couples numerical models with physical testing to evaluate structural performance under realistic loading conditions. In RTHS, critical components of a structure (e.g., dampers, isolators, or connections) are tested physically, while the rest of the structure is modeled computationally, and both are coupled in real time through control systems. Unlike conventional single-axis setups, multi-axis RTHS can reproduce complex three-dimensional dynamic loads, which allows engineers to practically capture nonlinear behaviors with reduced cost compared to full-scale testing.
Real-Time Hybrid Simulation and Seismic Analysis for Floating Structure
A 6-DOF real-time hybrid wind tunnel testing framework is developed to cross-validate coupled wind–wave response analyses and evaluate the performance of innovative floating bridge concepts. Seismic analysis methods and design guidelines are established for floating bodies to ensure safer and more reliable offshore infrastructure.
Elastic Actuation, Sensing and Control
This research develops a real-time force/torque control framework using elastic actuation, integrating online state estimation with adaptive compensation. Targeted at applications where the system model is difficult to obtain or evolves over time, the framework continuously adapts without relying on a predefined model. A dedicated elastic structural element is designed to serve as both a compliant actuation component and a precise force transducer, enabling accurate force feedback and high-fidelity force control under varying conditions. Ultimately, the framework bridges the physical and numerical environments in real time, ensuring delay-free bidirectional interaction.
Deep Learning Technologies for Assessment of Seismic Responses and Damage of Nuclear Power Plant Structures and Equipment
This study uses deep learning to predict seismic responses and assess damage in nuclear power plant structures and equipment. Hybrid simulation and experimental data from our lab are used for training and validation. The framework integrates response prediction, damage detection and classification, and damage severity and localization.
Vision-Based Structural Health Monitoring
Among various vision-based SHM approaches, our group focuses on UAV-borne RGB imagery and LiDAR point clouds to quantify bridge displacement and damage-sensitive dynamics. We design multimodal fusion and digital-twin frameworks to deliver high-fidelity 3D reconstructions and integrity assessments. In particular, we employ Structure-from-Motion (SfM) and 3D design information to build accurate models, which are periodically updated with UAV displacement measurements. This continuous process enables systematic monitoring of structural robustness and proactive bridge safety evaluation.
3D Isolation Floor for Nuclear Power Plants
3D base isolation system is effective in protecting critical equipment from seismic and impact loads, but its practical application remains limited. This research aims to technically overcome these challenges by developing an advanced isolation floor system to enhance the seismic and impact-resistance performance of high-frequency sensitive electrical equipment in nuclear power plants.