论文标题
部分可观测时空混沌系统的无模型预测
Threshold voltage instability by charge trapping effects in the gate region of p-GaN HEMTs
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。
In this work, the threshold voltage instability of normally-off p-GaN high electron mobility transistors (HEMTs) has been investigated by monitoring the gate current density during device on-state. The origin of the gate current variations under stress has been ascribed to charge trapping occurring at the different interfaces in the metal/p-GaN/AlGaN/GaN system. In particular, depending on the stress bias level, electrons (VG < 6 V) or holes (VG > 6 V) are trapped, causing a positive or negative threshold voltage shift {DVTH, respectively. By monitoring the gate current variations at different temperatures, the activation energies associated to the electrons and holes trapping could be determined and correlated with the presence of nitrogen (electron traps) or gallium (hole traps) vacancies. Moreover, the electrical measurements suggested the generation of a new electron-trap upon long-time bias stress, associated to the creation of crystallographic dislocation-like defects extending across the different interfaces (p-GaN/AlGaN/GaN) of the gate stack.
