论文标题

集体动力学和平均场极限近似中的一般维克斯克模型的凯奇理论

Cauchy theory for general Vicsek models in collective dynamics and mean-field limit approximations

论文作者

Briant, Marc, Diez, Antoine, Merino-Aceituno, Sara

论文摘要

在本文中,我们在动力学层面上为圆环和整个空间提供了局部库奇理论。我们认为相当通用的相互作用内核,非线性粘度和非线性摩擦。特别是,我们包括归一化的内核,当颗粒的通量消失时,它们显示出奇异性。因此,就动力学方程式的凯奇理论而言,我们扩展到更通用的相互作用,并完成了Gamba等人在Gamba等人中启动的程序。 al。 (2016年)(作者认为没有发生奇异性),而在Figalli等人中。 al。 (2017年)(作者证明在均匀案例中不会发生奇异性)。此外,我们得出了明显的较低存在时间,以及一个全球存在标准,除其他情况外,它适用于非氮构和原始维克斯克问题的长期理论,而没有任何先验假设。在本文的第二部分中,我们还为近似粒子极限的平均场限制建立了一个近似(正则化)系统,该系统每当通量不消失时与原始的系统相吻合。基于极限动力学方程的结果,我们证明,在短时间内,该近似粒子系统的动力学与原始奇异动力学相吻合的概率趋向于许多粒子极限。

In this paper we provide a local Cauchy theory both on the torus and in the whole space for general Vicsek dynamics at the kinetic level. We consider rather general interaction kernels, nonlinear viscosity and nonlinear friction. Particularly, we include normalised kernels which display a singularity when the flux of particles vanishes. Thus, in terms of the Cauchy theory for the kinetic equation, we extend to more general interactions and complete the program initiated in Gamba et. al. (2016) (where the authors assume that the singularity does not take place) and in Figalli et. al. (2017) (where the authors prove that the singularity does not happen in the space homogeneous case). Moreover, we derive an explicit lower time of existence as well as a global existence criterion that is applicable, among other cases, to obtain a long time theory for non-renormalised kernels and for the original Vicsek problem without any a priori assumptions. On the second part of the paper, we also establish the mean-field limit in the large particle limit for an approximated (regularized) system that coincides with the original one whenever the flux does not vanish. Based on the results proved for the limit kinetic equation, we prove that for short times, the probability that the dynamics of this approximated particle system coincides with the original singular dynamics tends to one in the many particle limit.

扫码加入交流群

加入微信交流群

微信交流群二维码

扫码加入学术交流群,获取更多资源