Computation of Adiabatic and Diabatic Potential Energy

Abstract

This study delves into quantum simulation, exploring its foundational principles and potential applications. On the one hand, the classic methods of energy computation were used. On the other hand, the link between molecular dynamics (MD) and machine learning (ML) was also investigated. In the first part dedicated to adiabatic and diabatic potential energy curve (PEC), we examine quantum simulation techniques, including diabatization using neural networks to capture complex molecular dynamics. For the KFr molecule, we present novel adiabatic and diabatic potential energy curves. In the second part related to molecular dynamics simulation based on machine learning, the link between MD and ML was highlight to overcome difficulties of the classic simulation methods. This research aims to develop new approaches to molecular dynamics beyond the Born-Oppenheimer approximation and to explore the potential of machine learning, particularly interpretable neural networks, for efficient and interpretable simulations.