论文标题
部分可观测时空混沌系统的无模型预测
Electromagnetic Force and Momentum in Classical Macroscopic Dipolar Media
论文作者
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。
Using realistic classical models of microscopic electric-charge electric dipoles and electric-current (Amperian) magnetic dipoles, it is proven that the Einstein-Laub macroscopic electromagnetic force on a macroscopic-continuum volume of these classical dipoles equals the sum of the microscopic electromagnetic forces on the discrete classical dipoles in that volume. The internal (hidden) momentum of the discrete Amperian magnetic dipoles is rigorously derived and properly included in the determination of the macroscopic force from the spatial averaging of the microscopic forces. Consequently, the Abraham/Einstein-Laub rather than the Minkowski macroscopic electromagnetic-field momentum density gives the total microscopic electromagnetic-field momentum in that volume. The kinetic momentum is found for the volume of the macroscopic continuum from Newton's relativistic equation of motion. It is shown that the difference between the kinetic and canonical momenta in a volume of the macroscopic continuum is equal to the sum of the "hidden electromagnetic momenta" within the electric-current magnetic dipoles and within hypothetical magnetic-current electric dipoles replacing the electric-charge electric dipoles in the classical macroscopic continuum. To obtain the correct unambiguous value of the force on a volume inside the continuum from the force-momentum expression, it is mandatory that the surface of that volume be hypothetically separated from the rest of the continuum by a thin free-space shell. Two definitive experiments performed in the past with time varying fields and forces are shown to conclusively confirm the Einstein-Laub/Abraham formulation over the Minkowski formulation.