论文标题
部分可观测时空混沌系统的无模型预测
Simulation of Graphene Nanoplatelets for NO$_{2}$ and CO Gas Sensing at Room Temperature
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。
This work reports the modeling and simulation of gas sensors made from pristine graphene nanoplatelets (P-GnPs) using COMSOL Multiphysics software. The mass balance equation was solved while including contributions of electromigration flux. An example GnP-based gas sensor was simulated to undergo exposure to NO2 and CO gases at different concentrations to understand the effects of adsorption. Various electrical properties and the overall sensor responses were also studied as a function of gas concentration in order to determine how viable such sensors could be for target gases. The results herein show that the resistance of the P-GnP-based gas sensor decreases when exposed to NO2 gas whereas an opposite trend is seen when CO gas is used for exposures, ultimately suggesting that the P-GnPs exhibit p-type behavior. Sensitivities of 23 % and 60 % were achieved when the P-GnP-based gas sensor was exposed to 10 mol/m3 concentration of NO2 and CO at room temperature, respectively. The data heavily suggest that a higher sensitivity towards CO may be observed in future sensors. These simulations will benefit research efforts by providing a method for predicting the behavior of GnP-based gas sensors.
