论文标题
部分可观测时空混沌系统的无模型预测
First experimental results of the spatial resolution of RSD pad arrays read out with a 16-ch board
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。
Resistive Silicon Detectors (RSD, also known as AC-LGAD) are innovative silicon sensors, based on the LGAD technology, characterized by a continuous gain layer that spreads across the whole sensor active area. RSDs are very promising tracking detectors, thanks to the combination of the built-in signal sharing with the internal charge multiplication, which allows large signals to be seen over multiple read-out channels. This work presents the first experimental results obtained from a 3$\times$4 array with 200~\mum~pitch, coming from the RSD2 production manufactured by FBK, read out with a 16-ch digitizer. A machine learning model has been trained, with experimental data taken with a precise TCT laser setup, and then used to predict the laser shot positions, finding a spatial resolution of $\sim$~5.5~\mum.
