论文标题

从Navier-Stokes千年提出问题到软件计算

From Navier-Stokes millennium-prize problem to soft matter computing

论文作者

Sharma, Saksham, Marcucci, Giulia

论文摘要

克莱数学研究所在2000年制定了七个未解决的数学问题的列表,这些问题可能会影响和指导21世纪的研究过程。 Navier-Stokes规律性问题是七个问题之一,迄今尚未解决。这些方程控制粘性液体的运动,例如液体,气体,凝胶,聚合物。本文是关于NSE问题的关注点:对过去和当前解决它的努力的回顾以及我们对这些方程式的不断变化的观点和理解。主要的辩论是要么证明Navier-Stokes方程(NSE)的解决方案是平滑的,要么证明解决方案在有限的时间内达到了奇异性(爆炸);给定一些初始条件。在过去,重点是寻找平滑解决方案的界限,但最近,人们对寻找爆炸的解决方案越来越兴趣。当前的文章通过讨论最近指出的是应将流体方程式视为“计算机程序”的想法,将Euler方程(NSE的子类)的Turing完整性置于中心阶段。鉴于这种情况,这篇文章随后主张从复杂性概念中看到“连续流体”,因此是图灵机。这一概念将流体力学的学科连接到诸如“神经形态计算,储层计算”之类的学科,并有可能导致新的学科,“软物质复杂性,计算和学习”,每个学科都简要介绍了。

Clay Mathematical Institute, in the year 2000, formulated a list of seven unsolved mathematical problems which might influence and direct the course of the research in the 21st century. Navier-Stokes regularity problem is one of the seven problems and has not been solved till date. These equations govern the motion of the viscous fluids, such as liquids, gases, gels, polymers. The present article is a spotlight on the NSE problem: a review of the past and current efforts to solve it, and our changing perspectives and understanding of these equations. The major debate is to either prove that the solutions to Navier-Stokes equations (NSE) are smooth or to prove that the solutions reach singularity at a finite-time (blowup); given some initial conditions. While in the past, the focus was on finding bounds for the smooth solutions, recently, there has been growing interest in finding solutions that blowup. The current article places the Turing-completeness of Euler equations (a subclass of NSE) into the centre-stage, by discussing the recent developments that are pointing towards that the idea that the fluid equations should be viewed as a 'computer program'. Given this is the case, the article then argues to see 'continuum fluid' from a complexity notion, and thus, as a Turing machine. This notion connects the discipline of fluid mechanics to disciplines like, 'neuromorphic computing, reservoir computing' and potentially leading to emerging new discipline, "soft matter complexity, computing, and learning", each of which has been discussed in brief.

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