Research Agenda

Research

My research focuses on decision-support-oriented multi-level intelligent condition monitoring of wind turbines, connecting turbine-level operational state representation, subsystem-level heterogeneous monitoring, critical component degradation monitoring, blade crack propagation prediction, and predictive maintenance effectiveness evaluation.

Multi-level Wind Turbine Condition Monitoring

This direction studies how to represent and monitor wind turbine operating states across turbine, subsystem, and component levels using SCADA data, vibration signals, simulation results, and machine learning models.

Gearbox and Bearing Fault Diagnosis

This direction investigates self-supervised learning, graph neural networks, transfer learning, and health-index construction for gearbox condition monitoring and bearing-related degradation analysis.

Blade Crack Propagation Prediction

This direction develops OpenFAST-Abaqus coupled simulation workflows and damage-index regression methods for large wind turbine blade crack monitoring, stiffness degradation analysis, and crack propagation prediction.

Predictive Maintenance and O&M Decision Support

This direction links monitoring outputs with maintenance effectiveness evaluation, risk assessment, and predictive maintenance strategy optimization.