论文标题
部分可观测时空混沌系统的无模型预测
Neutrino Mass Ordering -- Circumventing the Challenges using Synergy between T2HK and JUNO
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。
One of the major open problems of neutrino physics is MO (mass ordering). We discuss the prospects of measuring MO with two under-construction experiments T2HK and JUNO. JUNO alone is expected to measure MO with greater than $3σ$ significance as long as certain experimental challenges are met. In particular, JUNO needs better than 3$\%$ energy resolution for MO measurement. On the other hand, T2HK has rather poor prospects at measuring the MO, especially for certain ranges of the CP violating parameter $δ_{\rm CP}$, posing a major drawback for T2HK. In this letter we show that the synergy between JUNO and T2HK will bring two-fold advantage. Firstly, the synergy between the two experiments helps us determine the MO at a very high significance. With the baseline set-up of the two experiments, we have a greater than $9σ$ determination of the MO for all values of $δ_{\rm CP}$. Secondly, the synergy also allows us to relax the constraints on the two experiments. We show that JUNO, could perform extremely well even for energy resolution of 5$\%$, while for T2HK the MO problem with "bad" values of $δ_{\rm CP}$ goes away. The MO sensitivity for the combined analysis is expected to be greater than $6σ$ for all values of $δ_{\rm CP}$ and with just 5$\%$ energy resolution for JUNO.