论文标题
部分可观测时空混沌系统的无模型预测
W entropy in hard-core system
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。
As predicted by the second law of thermodynamics, the increase of entropy is irreversible in time. However, in quantum mechanics the evolution of quantum states is symmetrical about time-reversal, resulting a contradiction between thermodynamic entropy and quantum entropy. We study the W entropy, which is calculated from the probability distribution of the wave function on Wannier basis, in hard-core boson system. We find that W entropy and F entropy, which is calculated from the probability distribution of the wave function on Fock basis, satisfy an approximately linear relationship and have the same trend. Then, we investigate the evolution of W entropy for various parameters. We calculate the regression period of W entropy and find its dependence on the lattice scale. Our results show that the second law of thermodynamics is not completely valid in quantum mechanics. The behaviour of W entropy obeys the second law of thermodynamics, only when the system scale is large enough.
