论文标题
部分可观测时空混沌系统的无模型预测
Manipulating electron waves in graphene using carbon nanotube gating
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。
Graphene with its dispersion relation resembling that of photons offers ample opportunities for applications in electron optics. The spacial variation of carrier density by external gates can be used to create electron waveguides, in analogy to optical fiber, with additional confinement of the carriers in bipolar junctions leading to the formation of few transverse guiding modes. We show that waveguides created by gating graphene with carbon nanotubes (CNTs) allow obtaining sharp conductance plateaus, and propose applications in the Aharonov-Bohm and two-path interferometers, and a pointlike source for injection of carriers in graphene. Other applications can be extended to Bernal-stacked or twisted bilayer graphene or two-dimensional electron gas. Thanks to their versatility, CNT-induced waveguides open various possibilities for electron manipulation in graphene-based devices.
