Energy-Conserving Particle-Pushing Algorithms for Hybrid Fluid-Kinetic Simulations

TriForce is a computational environment using a hybrid fluid-kinetic model to execute higher-fidelity simulations in shorter time frames. At its core are a particle-in-cell model and a meshless hydrodynamic model that can be coupled to perform modeling across multiple spatiotemporal scales and approximation regimes. The key to performing these calculations over large numbers of time steps is the management of accumulating numerical error. Energy-conserving algorithms for the integration of the particle equation of motion, also called "particle pushers," are needed to maintain accuracy while simultaneously loosening the spatial and temporal resolution requirements of the simulation. However, no one algorithm is suited for every task. As such, the TriForce Fundamental Algorithm Testing Environment is being developed as a stand-alone platform in Python to provide the opportunity to quickly implement, characterize, and compare algorithms for further use in TriForce's Library for Integrated Numerical Kinetics, which is the kinetic half of TriForce. Presented here are initial, side-by-side analyses of particle-pushing algorithms for TFLink, comparing their performance across multiple computational and physical situations.