Dr. Kai-Qi LI

Postdoctoral Fellow

The Hong Kong Polytechnic University

My research program focuses on the intersection of Artificial Intelligence (AI) and Geotechnical Engineering, with a specific emphasis on developing Physics-Informed Machine Learning (PIML) frameworks to address complex challenges in geo-engineering. My work aims to bridge the gap between traditional numerical methods (e.g., FEM) and data-driven intelligence to achieve more reliable, efficient, and sustainable infrastructure design. My core research pillars include:

• Intelligent Geotechnical Modeling: Developing physics-informed neural networks (PINNs) and hybrid AI-FEM frameworks for constitutive modeling and multi-physics simulations (THMC coupling), particularly for frozen soils and hydrate-bearing sediments.
• Uncertainty Quantification & Reliability: Integrating stochastic modeling with deep learning to quantify and propagate uncertainties in complex geological environments, ensuring robust risk assessment for large-scale infrastructure projects.
• Digital Twins for Coastal & Underground Resilience: Leveraging multi-fidelity data and digital twin technology to monitor and predict the performance of critical infrastructure under extreme climate conditions and complex geological constraints.
• Multiscale & Multiphysics Analysis: Investigating the scale effects and meso-mechanical behaviors of geomaterials to provide a theoretical foundation for macro-scale engineering applications.