Research

My research in structural dynamics and vibration control focuses on mitigating the response of engineering structures subjected to dynamic environmental and operational loads. A major emphasis of this work has been the development of passive vibration-control strategies for offshore wind turbines, with particular attention to tuned mass damper systems and multidirectional motion control.

This work includes vibration reduction of fixed-bottom offshore wind turbines and motion mitigation of floating wind turbines through advanced tuned mass damper concepts designed to reduce structural response, fatigue demand, and platform motion.

Vibration Reduction of Fixed-Bottom Offshore Wind Turbines:

This research investigated the use of a pendulum tuned mass damper to reduce wind-turbine tower vibrations in both fore-aft and side-to-side directions. The proposed system was designed to provide multidirectional vibration mitigation while improving fatigue performance. Numerical simulations demonstrated substantial reductions in structural response and fatigue damage compared with conventional control configurations.
Motion Mitigation of Floating Wind Turbines:

This research focused on controlling the coupled motions of floating offshore wind turbines in pitch, roll, and heave. A three-dimensional nonlinear tuned mass damper was developed to provide simultaneous motion mitigation in multiple directions. The nonlinear characteristics of the system were investigated and optimized to improve its effectiveness under combined wind and wave loading.

                                  

Physical testing of offshore wind turbines is challenging because aerodynamic and hydrodynamic loads follow different scaling laws, making it difficult to simultaneously reproduce wind- and wave-induced effects in reduced-scale experiments. Real-time hybrid simulation provides an alternative by combining numerical simulation with physical testing, allowing different components of the system to be represented using the approach best suited to their behavior.

My research in this area focused on developing real-time hybrid simulation frameworks for offshore wind turbines subjected to combined wind and wave loading. The system is divided into numerical and physical substructures that interact in real time through actuators and sensors. The developed frameworks were used to investigate the accuracy and feasibility of hybrid testing, including the effects of actuator delay, measurement noise, and scaling errors on the predicted structural response.

 

Structural Health Monitoring and Digital Twins of Bridges
My research in structural health monitoring and digital twins for bridges aims to move condition assessment beyond periodic visual inspection toward continuous, data-informed evaluation. The work integrates field sensing, structural dynamics, finite-element modeling, system identification, and data analytics to detect changes in structural behavior and support more timely infrastructure assessment.
 
A major focus is the development of an AI-infused, physics-based digital twin that combines a calibrated finite-element model with measurement data and machine-learning methods. Environmental and operational effects, such as temperature and traffic variability, are separated from changes associated with the structure itself. Recursive estimation methods are then used to update physical parameters such as stiffness and damping, support damage detection, and enable virtual sensing of response quantities at locations where physical sensors are unavailable.
Field-scale monitoring is supported by sensors such as accelerometers, strain gauges, temperature sensors, and other structural and environmental instrumentation. These measurements provide the data needed for modal analysis, long-term condition tracking, and integration with digital-twin models for real-world bridge monitoring.
 
 
Condition Monitoring and Digital Twins of Drivetrain Systems
 
My research in drivetrain and gearbox condition monitoring focuses on combining physics-based modeling, sensing, and data analytics to understand machine behavior and identify developing faults. Vibration, speed, torque, temperature, and other measured responses can provide early indications of changes in gears, bearings, shafts, and other drivetrain components. This work investigates how measurements from healthy and damaged systems can be used for fault detection, anomaly identification, and predictive condition assessment under varying operating conditions.
 
A complementary direction focuses on digital twins and inverse modeling of wind turbine drivetrains. Physics-based drivetrain models are combined with measured responses using model-updating and Bayesian data-assimilation techniques to estimate uncertain system parameters and quantities that are difficult to measure directly, including aerodynamic loads. These approaches provide a framework for model validation, virtual sensing, condition monitoring, and digital-twin-based assessment of drivetrain performance.
 
Experimental investigations also consider common gearbox and bearing damage scenarios including cracked, chipped, worn, and missing gear teeth and bearing defects, using multi-sensor measurements to study how faults appear in the dynamic response of rotating machinery.
 
 Structural Damage Identification of Offshore Wind Turbines:
My research in offshore wind turbine damage identification focuses on combining physics-based numerical models with measured structural responses to detect, locate, and quantify changes associated with damage. Finite-element model updating and system-identification techniques are used to track changes in structural properties and translate measured response variations into physically meaningful indicators of structural condition.
 
A two-step damage-identification framework was developed for offshore wind turbine structures. In the first step, a global finite-element model is updated to identify the structural component affected by damage. A refined local model is then used to determine the location and severity of the damage within the identified component. This multi-scale strategy provides a computationally efficient approach for progressing from system-level detection to more detailed damage assessment. Related work has also investigated damage detection in floating wind turbine mooring systems using response-based signal processing and classification methods, extending condition-monitoring concepts beyond the turbine structure to other critical components of offshore wind energy systems.