Grid induced noise and entropy growth in 3d particle-in-cell simulation of high intensity beams

TL;DR

This paper investigates the numerical noise in 3D particle-in-cell simulations of high intensity beams, confirming theoretical predictions and revealing additional grid-related noise effects, with implications for high current linac modeling.

Contribution

The study validates the analytical model of entropy growth due to temperature anisotropy and uncovers a new grid-induced noise mechanism affecting simulation accuracy.

Findings

Entropy growth depends on bunch temperature anisotropy as predicted.

A non-Liouvillean noise mechanism from the Poisson solver grid is identified.

Results aid in balancing particle number and grid resolution in simulations.

Abstract

The numerical noise inherent to particle-in-cell simulation of 3D high intensity bunched beams is studied with the TRACEWIN code and compared with an analytical model by Struckmeier. The latter assumes that entropy growth can be related to Markov type stochastic processes due to temperature anisotropy and the artificial collisions caused by using macro-particles and calculating the space charge effect. The resulting noise can lead to growth of the six-dimensional rms emittance and a suitably defined rms entropy. We confirm the dependence of this growth on the bunch temperature anisotropy as predicted by Struckmeier. However,we also find an apparently modified mechanism of noise generation by the non-Liouvillean effect of the Poisson solver grid, which exists in periodic focusing systems even in the complete absence of temperature anisotropy. Our findings are applicable in particular to high current linac simulation, where they can provide guidance to estimate noise effects and help finding an effective balance between the number of simulation particles and the grid resolution.