DEM2GATE: Combining discrete element method simulation with virtual positron emission particle tracking experiments

Abstract

Positron Emission Particle Tracking (PEPT) is a Lagrangian particle tracking technique useful for imaging flow in opaque granular media. The scientific and industrial systems studied using PEPT and the PEPT technique can also be simulated: Discrete Element Method (DEM) for granular systems and Monte Carlo for PEPT. Using the open-source DEM2GATE library, we quantify the reconstruction of particle trajectories and overall system behaviour by integrating particle trajectories from a rotating drum simulation with a PEPT detector simulation. The results show the main drivers of PEPT spatial error are particle acceleration and gamma-ray scattering. The accuracy of reconstructed flow fields increases logarithmically with measurement time. Measurements of higher-order, derived quantities require more time to reach a desired level of accuracy. Surprisingly, our results imply PEPT measurements relying on fluctuating velocity deviate substantially from the system's true behaviour. However, using DEM2GATE, these errors are predictable and can be corrected.