Field Sampling Strategies for POD Model Order Reduction of Particle-in-Cell Simulations

Particle-in-Cell (PIC) simulations are an useful tool for the analysis of kinetic collisionless plasmas such as those found in high power microwave devices, vacuum electronics, directed energy devices, among others. However, PIC simulations remain computationally costly, and model order reduction strategies such as the Proper Orthogonal Decomposition (POD) can be instrumental in reducing some of these costs. The efficacy of the reduced-order model resulting from application of POD depends on several factors, among them the strategy employed to sample field values when constructing the reduced order model. In this work we analyze and compare several such field sampling strategies.