摘要

We establish an optimal error estimate for a random particle blob method for the Keller-Segel equation in R-d (d >= 2). With a blob size epsilon = h(kappa) (1/2 < kappa < 1), we prove a rate h| ln h| of convergence in l(h)(p) (p > d/1-kappa) norm up to a probability 1 - h(C| ln h|), where h is the initial grid size.