Learning heterogenous reaction rates from stochastic simulations

Reaction rate equations are ordinary differential that frequently used to describe deterministic chemical kinetics at the macroscopic scale. At microscopic scale, is stochastic and can be captured by complex dynamical systems reproducing spatial movements of molecules their collisions. Such molecular dynamics may implicitly capture intricate phenomena affect reaction rates but not accounted for in models. In this work we present a data assimilation procedure learning nonhomogeneous kinetic parameters from simulations with many simultaneously reacting species. The learned then plugged into predict long time evolution system. way, our discovers an effective equation kinetics. To demonstrate procedure, upscale system forms covalently bonded network severely interfering rates. Incidentally, report feature peculiar temperature dependences, whereas probability strand close cycle follows universal distribution.