反应势发展面临的挑战及其在含能材料模拟中的研究进展OA

Focusing on the recent developments in reaction potential functions and their applications and challenges in the simu-lation of energetic materials,the impact of key factors such as the reaction potential function models and their modifications,charge correction schemes,force field development strategies,and training-set construction were systematically discussed.In addition,recent advances in machine-learning-based reactive potentials and coarse-grained models for reactive simulations were also highlighted.Despite these developments,existing approaches still face challenges in balancing accuracy and compu-tational efficiency,ensuring parameter transferability,description of electronic-structure effects,and adequately sampling criti-cal regions of the reaction landscape.To address these challenges,microscopic simulation of energetic materials is accelerating towards the integration of data-driven approaches with physical principles.This trend is primarily reflected in the construction of high-accuracy and high-efficiency machine-learning potentials,the development of automated and intelligent parameter optimi-zation strategies,and the establishment of multi-scale simulation frameworks.In the future,by further integrating physical laws with data-driven methodologies,developing intelligent potential functions with enhanced transferability and predictive capabili-ty,and constructing unified multi-scale simulation platforms,it is expected to enable more accurate descriptions of complex re-action mechanisms of energetic materials and provide more reliable theoretical foundations for their rational design,perform-ance optimization,and engineering applications.123 references were attached.