论文标题

部分可观测时空混沌系统的无模型预测

Vacuum-dual static perfect fluid obeying $p=-(n-3)ρ/(n+1)$ in $n(\ge 4)$ dimensions

论文作者

Maeda, Hideki

论文摘要

储层计算是预测湍流的有力工具,其简单的架构具有处理大型系统的计算效率。然而,其实现通常需要完整的状态向量测量和系统非线性知识。我们使用非线性投影函数将系统测量扩展到高维空间,然后将其输入到储层中以获得预测。我们展示了这种储层计算网络在时空混沌系统上的应用,该系统模拟了湍流的若干特征。我们表明,使用径向基函数作为非线性投影器,即使只有部分观测并且不知道控制方程,也能稳健地捕捉复杂的系统非线性。最后,我们表明,当测量稀疏、不完整且带有噪声,甚至控制方程变得不准确时,我们的网络仍然可以产生相当准确的预测,从而为实际湍流系统的无模型预测铺平了道路。

We obtain the general $n(\ge 4)$-dimensional static solution with an $(n-2)$-dimensional Einstein base manifold for a perfect fluid obeying a linear equation of state $p=-(n-3)ρ/(n+1)$. It is a generalization of Semiz's four-dimensional general solution with spherical symmetry and consists of two different classes. Through the Buchdahl transformation, the class-I and class-II solutions are dual to the topological Schwarzschild-Tangherlini-(A)dS solution and one of the $Λ$-vacuum direct-product solutions, respectively. While the metric of the spherically symmetric class-I solution is $C^\infty$ at the Killing horizon for $n=4$ and $5$, it is $C^1$ for $n\ge 6$ and then the Killing horizon turns to be a parallelly propagated curvature singularity. For $n=4$ and $5$, the spherically symmetric class-I solution can be attached to the Schwarzschild-Tangherlini vacuum black hole with the same value of the mass parameter at the Killing horizon in a regular manner, namely without a lightlike massive thin-shell. This construction allows new configurations of an asymptotically (locally) flat black hole to emerge. If a static perfect fluid hovers outside a vacuum black hole, its energy density is negative. In contrast, if the dynamical region inside the event horizon of a vacuum black hole is replaced by the class-I solution, the corresponding matter field is an anisotropic fluid and may satisfy the null and strong energy conditions. While the latter configuration always involves a spacelike singularity inside the horizon for $n=4$, it becomes a non-singular black hole of the big-bounce type for $n=5$ if the ADM mass is larger than a critical value.

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