{"id":"arxiv/2608.16294","url":"https://arxiv.org/abs/2608.16294","tags":["math.NA"],"score":0,"title":"Hierarchical sparse-grid particle-in-cell method with locally adaptive mesh refinement","domain":"math.NA","source":"arxiv","authors":["F. Deluzet","C. Guillet","J. Narski","P. Pace"],"reviews":0,"subject":"math.NA","summary":"In this paper, we introduce new approximation spaces and a locally adaptive refinement strategy for the hierarchical sparse-grid PIC (HSG-PIC) method to improve the bias while preserving the noise-reduction properties of sparse-grid methods. We first propose an energy-based approximation space, which optimizes the relation between the $\\mathrm{H}^1$-norm error and the number of degrees of freedom, together with a family of generalized sparse-grid spaces that continuously connects classical sparse-grid and full-grid approximations. We then develop a locally adaptive approximation strategy…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-08-17","reactions":{}}