Research Interests

Machine learning for OPF

Predicting solutions for optimal power flow (OPF) problems by deep neural networks.


Developing and implementing efficient preconditioners for geometry optimisation and transition state search of molecular and material systems.

Machine learning for PMF

Introducing a potential of mean force (PMF) technique by reconstructing high dimensional free energy surfaces from instantaneous collective forces (ICF) using Gaussian process regression (GPR).

GAP for organic molecules

Developing machine learning based Gaussian approximation potentials (GAP) for organic molecules using high level ab initio calculations.

Collective variables

Introducing highly efficient collective variables for evaluating free energy profiles / surfaces of chemical reactions in complex systems.

Hybrid QM/MM methods

Developing, implementing and testing hybrid quantum mechanical - molecular mechanical (QM/MM) techniques. Introducing the adaptive buffered-force QM/MM method.

In silico enzyme design

Studying the origin of enzymatic catalytic effect. Using reorganization energy in rational enzyme design.

Enzymatic reactions

Investigating and understanding chemical reactions in enzymatic environments using molecular dynamics (MD) simulations.

Solvated electron

Investigating and understanding solvation dynamics of electron in bulk methanol and methanol clusters.