8-10 June 2016
Asia/Bangkok timezone

Investigating rare events with modified Monte Carlo method based on fluctuation theorem

Jun 8, 2016, 2:15 PM
15m
Room E1

Room E1

Oral presentaion Statistical and Theoretical Physics Session XI

Speaker

Mr Takol Tangphati (Department of physics, Chulalongkorn university)

Description

The Monte Carlo (MC) method is a useful tool to solve a broad range of problems numerically; random numbers are used to decide whether a process is accepted resulting in a desired equilibrium distribution. In certain complex problems where transitions to some states are rare, however, this method dramatically consumes more resources and spends a long time trying to find the equilibrium states. To improve upon the conventional MC, we apply a fluctuation theorem (FT) to MC which provides a relative weight of a backward trajectory with respect to the forward one. Interestingly, not only does FT double information but also assists a probability of system to converge to equilibrium more rapidly. We apply the algorithm to solve a test problem where the energy barrier to a certain state is comparatively high. The result of the simulation is consistent with the analytic solution, but with significant speed gain compared to that from the conventional method.

Primary author

Mr Takol Tangphati (Department of physics, Chulalongkorn university)

Presentation materials

There are no materials yet.