Research

Research Overview:

Our group is broadly interested in computational mechanics and artificial intelligence for applications related to energy, sustainability and resilience of the infrastructure and the built environment. In particular, we are interested in energy and infrastructure sustainability, climate-resilient civil infrastructure, climate change mitigation technologies, and materials related to energy and civil infrastructure. Some of the applications of our work include scientific machine learning, clean/renewable energy and decarbonization technologies (e.g., carbon sequestration, geothermal energy production, hydrogen storage), natural hazards (e.g., earthquakes and landslides), climate change adaptation, analysis and design of composite materials, multiscale material data fusion and advanced infrastructure materials and systems.

If you are interested in any of these topics or would like to discuss new ideas, check the information under Join us, and/or contact me via email.  

Research Areas:

Multi-scale multi-physics modeling of complex materials.

Our research includes characterization and modeling of heterogeneous materials (e.g. rocks, soils, cementitious materials, and composites) across scales, and development of multi-scale and multi-physics models to link the microstructure and macroscopic behavior of these materials. For this purpose, we combine experiments, mechanics, computational methods, and statistical and machine learning techniques.

Microstructure of synthetic and natural materials determine their overall properties and performance in engineering applications. We develop advanced techniques to incorporate microstructure of materials into computational models for applications such as subsurface energy systems and composite materials and structures. 

Scientific machine learning and physics-guided artificial intelligence.

In this research area, we develop novel computational methods by enhancing computational mechanics tools with machine learning techniques. In particular, we are interested in developing fast and efficient computational techniques for modeling of complex systems and digital twin applications.

Multi-scale material characterization and data fusion via statistical and machine-learning techniques. 

We integrate advanced material characterization tools with statistical and machine learning techniques to probe the material structure and properties across scales to enhance analysis and design of materials.

Natural hazards and risk assessment:

We develop machine learning and computational mechanics techniques to model natural hazards (for example, landslides, floods, storms, wildfires, and earthquakes) and assess their risks and impacts on civil infrastructure.

Renewable energy and decarbonization technologies:

We address challenges in subsurface renewable energy technologies such as carbon sequestration, enhanced geothermal systems and hydrogen storage at both material level and system level. Examples include hydro-chemo-mechanical behavior of rocks, effects of rock-fluid interactions, and advanced computational techniques for large-scale modeling of subsurface systems.